research and generate ideas

The Writing Process

Making expository writing less stressful, more efficient, and more enlightening

Step 1: Generate Ideas

“If at first the idea is not absurd, then there is no hope for it.” —Albert Einstein  

The first step of the writing process (that is, after carefully reading and understanding the assignment) is to generate ideas for your project. In shorter versions of the writing process, or in processes designed for other kinds of writing, step 1 is sometimes called “gathering” because it also includes doing research.

In expository writing, though, even for a research paper, you will want to “generate ideas” first. Why? First, you will want to see what you already know and think about a subject. Second, you will want to see what ideas you can come up with yourself. And third, the methods below will help you define what questions you want to start your research with.

There are various ways to generate ideas for your writing. People think and learn differently, so try them all and choose the one that’s best for you—although if you have never tried freewriting or “moodling,” described below, I strongly encourage you to try them both at least once. My experience is that students are usually happily surprised at the results.

Moreover, freewriting is often useful to non-native speakers of English who still struggle with fluency (i.e., writing quickly or relatively easily, in contrast to accuracy, which an overriding concern for at this stage of the writing process can inhibit the flow of words and ideas).

As you will recall from “ Whom are you writing for? “, the writing process starts as writer-oriented and gradually moves toward a reader-oriented product. Thus this step should be thought of as completely writer-oriented . Forget about your reader and assume that no one is going to see your notes or ideas from this stage. Let your ideas come out freely and be as wild and crazy as they seem. If you immediately censor what you think might be “dumb” or “silly” ideas, you may eliminate good ideas or connections to good ideas and are not really letting yourself think. As one moderately successful scientist (pictured at left) once put it, “Imagination is greater than knowledge. Knowledge is limited. Imagination encircles the world.”

There are four primary methods of generating ideas:

• Brainstorming
• Freewriting
• Idea Map/Web
• “Moodling”

Photo “Büro im Wasser” from Bundesarchiv, Bild 102-08112 / CC-BY-SA. Licensed under CC BY-SA 3.0 de via Wikimedia Commons.

Photo “Ideas” ©2014 Rafal Knop

Idea Generation Techniques for Research Projects

By Jin Chow

Co-founder of Polygence, Forbes 30 Under 30 for Education

11 minute read

Often, the first few ideas for a project are not the best. We can often miss hidden connections or overlook more exciting opportunities. Idea generation techniques can help you dig deeper, past the more obvious or cliched ideas, and into exciting new territory. In this post, we’ll go over the techniques of brainstorming, mind mapping, SCAMPER, random word, role reversal, visualization, and idea combinations, all powerful tools to stimulate creative thinking.

It’s important to harness the power of diverse idea generation if you want to come up with an innovative project. But descriptions of these techniques might not easily translate to your process. That’s why we’ll also give you specific examples of how other students have used these techniques for research in many different fields, such as comp sci, biology, product design, environmental activism, business, and history.

What is Brainstorming?

Brainstorming is such an effective way of generating ideas that we wrote a whole post about it. The basic idea of brainstorming is free-associating as many connected thoughts as you can while withholding any judgment. The goal here is quantity. Quality comes later. 

Brainstorming is particularly effective in groups because your combined efforts can produce more ideas faster than when you’re brainstorming alone. One person's idea can also trigger another person's thoughts and creativity, leading to a chain reaction that might not have emerged in an individual brainstorming session.

You can incorporate any of the other 6 techniques in this post in your brainstorming session, done alone or with others. Mind mapping is, in fact, just a visual form of brainstorming.

Steps for brainstorming

Define the problem you want to brainstorm. 

Create a quiet environment with the tools you need to capture thoughts quickly and accurately.

Set a time limit.

Defer judgment by adopting a “yes, and…” policy as they do in improv.

Aim for as many ideas as possible.

Pick out your most promising ideas only after the session is over.

Example: using brainstorming to generate a Computer Science project idea

Jessie isn’t sure what kind of project she wants to do, but she loves music and is intrigued by the possibilities of AI. That’s a good enough place to start. As her brainstorming question, she comes up with: How can AI help an aspect of musical creation, performance, or distribution? She writes this on top of a white board.

She gets together with a band friend and her brother who is a great programmer and knows a lot about AI. They agree to say any questions or ideas that come into their heads and jot them down on sticky notes. Nobody is allowed to shoot down any idea at this point. They give themselves 60 minutes to simply riff on ideas without worrying about how Jessie would actually get it done. Here are a few examples of things they write on stickies. 

Some are new questions, some are thoughts, some are topical issues. Some are related to each other and others aren’t. But the main thing is that they all address AI and music, and they help Jessie start to narrow in on a more specific idea that she can get really excited about and start to research further. 

What is Mind Mapping?

Mind maps are a great way to organize and represent connections between ideas visually. They can be used to help you organize complex topics when you’re studying for an exam, but they can also be great brainstorming tools. (You’ve probably also seen mind maps on bulletin boards in detective shows as they gather suspects, evidence, and clues.) Because they are non-linear, they can reveal hidden connections and help you develop more creative, unexpected ideas. Mind maps generally start with a central idea and then branch out to related thoughts, keywords, or connections. There are many great free online tools you can use now for mind mapping ( FigJam , Miro , and MindMeister come to mind), but good old fashioned paper and pencil works great too. 

Steps for mind mapping

At the center of your mind map, start with a basic topic, problem, or question.

Start to free-associate subtopics or solutions and connect them to the central topic with branches. 

Explore each subtopic and add more ideas around them.

Try to use concise words and/or images to keep the mind map from getting too convoluted.

Keep exploring connections. Allow yourself to freely associate and add ideas as they come to you.

Example: using mind mapping to generate a Biology project idea

Logan is interested in gut health and how it can impact your immune system as well as your mental health. He does a brainstorming session in which he mind-maps all of the different factors of the human microbiome that he knows about and that interest him. As he’s working on the mind map, he’s also looking things up online and adding new ideas and facts as he discovers them.

Here is his mind map (done in MindMeister ):

Looking at his mind map, Logan sees a correlation between his interest in microbial diversity and improved immune system, and begins to wonder if something as simple as eating probiotics such as yogurt, sauerkraut, or miso could help you fight viral infections like COVID-19 and other diseases. Ater filling out this mind map, he can connect specific areas of interest to each other and do a bit more research as he plans out a specific hypothesis and strategy for his project.

It’s a good idea to keep all of your original mind maps, even after you have started your project. You never know when you might have to pivot in a slightly different direction or want to try new combinations as you learn more about your topic. 

What is the SCAMPER Technique?

SCAMPER is an acronym for a set of thought experiment prompts you can use to come up with unexpected ideas. You can use it during a brainstorming or mind mapping session to get you moving in a different direction if you feel stuck.

What SCAMPER stands for

S ubstitute

P ut to another use 

E liminate 

R everse/Rearrange

Example: using the SCAMPER technique to generate a Product Design project idea

Shamir is interested in product design and helping refugees who have been seeking shelter and camping in his community. Here’s an example of how he used the SCAMPER technique to brainstorm various small products he could gather into a helpful kit. 

He starts with the basic question: "What are some products we can create or collect to help refugees with their daily struggles?”

This is what he comes up with:

Substitute:  First, he considers substituting something critical that they would need in daily life with something else. What if you substitute traditional cooking methods with portable and efficient solar-powered cookstoves? Or replace traditional water collection methods with lightweight, collapsible water containers for easier transport?

Combine: Next, he thinks about ways to combine different things they might need together into one thing. You could combine a wearable translation device with a language learning app to help refugees communicate and learn the local language. Or combine a basic mobile phone with a preloaded app for accessing important information about local services, healthcare, and education.

Adapt: He looks at items refugees already were using and wondered how he could adapt and evolve a better design to solve problems. A backpack could be adapted to include solar panels for charging electronic devices, ensuring refugees can stay connected even in remote areas. A traditional tent could be adapted to be more easily assembled and lightweight.

Modify: Similarly, he explores ways to modify or alter existing tools and products to make them even more useful and solve specific problems. A basic first aid kit should include instructions in multiple languages and contain culturally sensitive medical supplies. Simple clothing design could include built-in pockets for storing personal documents securely.

Put to another use: He also wonders how products out in the world could be repurposed in the context of the refugee shelter. Thinking on a larger scale, he realizes shipping containers could be repurposed into temporary modular shelters equipped with basic amenities. Recycled materials could be used to create low-cost, sturdy school desks for children.

Eliminate: He identifies some problematic elements that can be removed or eliminated without compromising daily life. The need for traditional and possibly dangerous cooking fuels could be eliminated by introducing solar-powered, portable food heaters. Again, language barriers could be broken down by developing a user-friendly, gesture-based communication app for refugees.

Reverse/Rearrange: Finally, Shamir tries reversing the roles of certain daily items and rearranges them to see if he could come up with new insights. A portable bed used at night could be reversed into a seating area to use during the day. A traditional grocery store could be rearranged into a pop-up marketplace offering more culturally familiar foods.

What is the Random Words Technique?

In random word brainstorming, you create fresh or disruptive ideas by associating a random word or phrase with a problem, question, or topic you want to explore. It's another form of brainstorming that helps you break past the more predictable and overused ideas most people come up with when first exploring a subject.

There are many ways you can come up with a random word: 

Open a dictionary to a random page and choose the first word you see.

Use online tools like Miro or Random Word Generator .

Turn on the radio or TV and write down the first few words you hear.

Write down words you hear throughout the week on scraps of paper and put them in a jar. Select one.

Then, take your random word and apply it to your area of interest. What thoughts are triggered by these two combined ideas?

Example: using the random word technique to generate an Environmental Studies project idea

Sophia wants to do a project involving carbon emissions in her school and propose strategies to reduce them, but she’s stuck. She uses an online random word generator tool. She tries applying the following word to her carbon emissions topic and comes up with the following ideas.

Random word: “Forget” 

Idle Forgetfulness: Parents who drive up to the school and are waiting for their kids often forget to turn their engines off. What if she were to monitor driver behavior in the school parking lots for a week after school and calculate how much carbon emission could be reduced simply by not idling? These findings could turn into a podcast about simple changes we can make in everyday life to dramatically reduce emissions if everyone followed suit. 

Carbon Forget-Me-Not Challenge: She could develop a social media challenge that encourages individuals to document and share all the ways they reduced their carbon footprint for a day so as to not forget their efforts, inspiring others to do the same.

Forget Carbon, Remember Bicycles: Create a project that encourages communities to "forget" about cars for a day or week. During this time, people would rely solely on bicycles for transportation. The project could include events like bike parades, awards for people who manage to go car-less for a full week, maintenance workshops, and education on the benefits of cycling.

What is the Role Reversal Technique?

Role reversal technique is another creative thought experiment that encourages you to view a problem or situation from different points of view by temporarily taking on the role of another person. It's an effective technique because it encourages you to step outside your usual thought patterns and consider viewpoints you might not have even identified or explored otherwise. For that reason, it is often used to break down mental barriers, stimulate new ideas, and promote empathy. It works best if you can work in a group so you can assign each person a role and then switch roles, but it can also work as a solo thinking endeavor. The role reversal technique is also very helpful for dealing with issues that involve diverse perspectives or complex human interactions.

How role reversal works

Define the problem or scenario.

Identify specific roles of the individuals or even inanimate objects associated with the issue.

Assign each person in your brainstorming session to one of these roles.

Each person comes up with ideas or insights from the point of view of the role they are playing.

Share and discuss all of these ideas.

Analyze the ideas as a group and identify which ones are the most promising or feasible.

Example: using role reversal to generate a Business project idea

Henry wants to set up a tech support business for the elderly people he knows who struggle to keep in touch. The problem is that he also knows they can’t afford expensive gadgets or tech services.

He goes through the role reversal process as follows:

Problem/Scenario: Elderly individuals struggle to keep in touch due to limited resources, visual impairment, and lack of tech savviness.

Multiple Perspectives to Explore:

The elderly individual on a limited budget - Doesn’t have the money or inclination to buy expensive new tech. Doesn’t understand how to use it even if it is gifted.

Younger caregiver or family member - Understands the challenges and responsibilities of caring for seniors who need tech support. They also want to be able to communicate more easily.

Tech enthusiast employee - Wants to be of assistance and bridge the digital divide and, of course, get paid.

Local community leader - Wants to promote local community initiatives that address the needs of elderly residents and get recognition for creating these initiatives.

Tech retailer or manufacturer - May want to explore partnerships or discounts in exchange for good press and free marketing. 

Social worker and/or healthcare provider - Understands the broader social, emotional, and health benefits of improved communication and tech access for the elderly. Work directly with seniors and have earned their trust.

Nonprofit organization representative - Might be able to use existing resources and senior programs to aid elderly individuals.

Henry combines the learnings from exploring each of these different roles, and comes up with a few tech solutions that might generate revenue while still relieving seniors from the financial burden of new tech and tech training.

Affordable Virtual Tech Workshops for Caregivers, Social Workers, and Healthcare Providers: Organize affordable tech workshops for family members and caregivers of the elderly, teaching them how to provide tech support. Charge a fee for workshop participation paid for by family members. This would not only generate revenue and empower seniors, but also empowers caregivers to assist their elderly family members more effectively.

Partnerships with Tech Retailers: Partner with tech retailers or manufacturers to offer discounts on tech products that are specifically useful for the elderly. The revenue would come from commissions or fees for promoting these deals. Seniors would gain access to affordable tech solutions, and the project would earn revenue through the partnership.

An Elderly-Friendly App/Device that Promotes Family Communication: Develop and sell a senior-friendly mobile app with a more senior-friendly device that helps them connect and laugh with a younger generation. This app/device could gamify the senior/youth interaction while ensuring more frequent communication and serving as a helpful health monitoring method for their caretakers. Seniors would gain access to more social involvement in their lives, and their family gets to know them better, while revenue comes from app sales and in-app purchases paid for by the younger generation and/or a software partnership.

How Does Visualization Work?

Visualization is another thought technique that helps you tap into your subconscious creativity by pushing you to explore your imagination in novel ways. The basic gist of it is that you do a deep dive into an imaginary situation by exploring all your sensory and emotional reactions. It can be used along with any of the other idea-generation strategies in this post. For instance, you can use it to really explore each point of view of each in the key roles you identify in the role reversal exercise. You can also find guided visualizations for creativity on YouTube and Audible.

How visualization works 

Define the problem or goal.

Find a quiet and relaxing environment in which to work. 

Close your eyes to block out distractions. Sit or lie down.

Imagine a concrete scenario related to your problem or goal.

Start to engage all of your senses. What do you hear, feel, smell, touch, and taste in this scene? The more specific and vivid the mental picture becomes, the more effective your visualization will be. 

You can write your own or use a pre-recorded guided imagery script designed to stimulate creativity. These can help you stay focused and provide a structured approach to visualization.

Stay relaxed and patient. It’s important not to rush ideas and trust that your subconscious will provide insights when you least expect it. Even if you don’t come up with a groundbreaking idea while you’re visualizing, it may come later as a result.

Capture your thoughts in writing after your session. Even if the ideas feel disconnected, they might lead to unexpected ideas.

Try to do a few sessions over a period of time and review the ideas you’ve written down.

Example: using visualization to generate a History project idea

Corita decides to use visualization to help her explore the unexpected roles women may have played in the American Civil War.

Here is how she goes about it and what she discovers along the way:

Setting the scene: Corita closes her eyes and imagined herself standing in the midst of a Civil War-era battlefield. She can see the soldiers in uniform, hear the distant sounds of cannon fire, and feel the tension in the air.

Engaging all the senses: In this visualization, she shifts her attention to the women who are present on the battlefield. Initially, she sees them as nurses or camp followers, but she tries to look beyond these well-known roles by using all her senses.

Sight: She notices the women tending to the wounded, but also observes women carrying messages, delivering supplies, or even fighting disguised as soldiers.

Sound: She listens to the conversations and interactions between women and soldiers, paying close attention to the women's determination and resourcefulness.

Touch: She feels the texture of the clothing women wore during this time, and imagines the physical challenges they faced wearing skirts as they navigated through the battlefield.

Corita also allows her mind to wander. She visualizes a woman taking on a leadership role, organizing a group of women to provide critical intelligence to the army. She imagines another woman who becomes an expert sharpshooter, secretly joining a regiment and making a significant impact on the battlefield. She pictures the women who created underground networks to support escaped slaves and Union soldiers.

She then dives into the emotions and experiences of these women. She can feel their determination, courage, and commitment to their causes and understands the risks they took and the challenges they overcame.

She takes notes on all the scenarios, characters, and insights that emerge during her visualization. These notes can serve as a starting point for her research. She might look for women sharpshooters and underground leaders. Or she might research the types of clothing women wore and if they adapted their attire in order to do their jobs more efficiently. The idea is that her imagination took her to surprising potential areas of interest.

What is Idea Combination?

Idea combination or idea synthesis is a creative thinking technique that involves merging or blending existing concepts or elements to generate new and innovative ideas. The smartphone, for instance, is an example of combining mobile phones, digital cameras, email, web browsers, GPS, and more into one handheld device. In chemistry, CRISPR-Cas9 gene editing came about as a combination of the discovery of CRISPER sequences, Cas Proteins, and prior gene editing techniques.  

Steps to start generating your own new hybrid solutions

Gather a diverse range of existing ideas, procedures, products, or concepts related to that problem or goal. These can come from research, observations, or other brainstorming sessions. Try to collect ideas from a wide range of different domains. The more diverse the ideas, the more unexpected the combination.

Deconstruct these ideas into their simplest parts.

Look for patterns, themes, or recurring elements that might appear across your collection.

Based on the commonalities you’ve identified, start brainstorming ways you can merge items from your list in creative ways, even if they seem unrelated at first. Think about how these combinations might lead to new and unique concepts or solutions.

Example using idea combination to generate a Medical Research project idea

Ryan is interested in developing a research project around a more well-rounded medical management plan for Type 1 Diabetes.

He uses the idea combination method to help himself expand the ideas that initially come to mind. 

Identify key elements and concepts: He begins by exploring various medicinal aspects, such as diagnosis, treatment, research, patient care, and healthcare systems for Type 1 Diabetes, so that he can clearly understand its causes, symptoms, management, and impact on individuals' lives.

Explore cross-disciplinary connections: He considers other fields or subjects that intersect with medicine and Type 1 diabetes, such as:

Biotechnology: He investigates innovations in insulin delivery or glucose monitoring technologies.

Nutrition and dietetics: He examines dietary strategies for managing blood sugar levels.

Psychology: He explores the emotional and psychological aspects of living with diabetes.

Brainstorm hybrid concepts: Next, he begins to merge his research from medicine, biotech, nutrition, and psychology to brainstorm a novel insulin delivery device that combines these insights.

Consider real-world applications: He investigates all of the existing insulin delivery devices and rates them to see how much they do or don’t address these hybrid concepts. He thinks about how these could be enhanced or a new solution developed that would address as many of the concerns he identified as possible. 

How Can Polygence Help With Idea Generation?

Some of the best ideas are refined through discussion, but it’s not always easy to find people able or willing to serve as a sounding board or as an additional expert source of information. You can use any or all of these techniques with a Polygence mentor. They can help provide expert feedback and guide you toward relevant information and resources. With either the Polygence Pathfinders or Launchpad Bundle , you can meet with three different mentors, float ideas with them, and get help defining your project ideas.

Final Thoughts

The freedom to come up with your very own project is exciting, but it can also be daunting. Brainstorming, mind mapping, SCAMPER, random word, role reversal, visualization, and idea combination are all great techniques for organizing your thoughts and revealing new directions and insights. 

If you want additional inspiration to get started, check out the types of research ideas that Polygence mentors have helped high school students complete. And here’s another post of passion project ideas . Or read about another technique that can help you narrow down your topic: the Japanese concept called ikigai. At Polygence, we’re here to help you unleash your creative potential and innovate on projects you can’t wait to work on.

Get Matched with a Mentor

Interested in doing one of these exciting research projects? Click below to get matched with one of our expert mentors!

A business journal from the Wharton School of the University of Pennsylvania

Generating Ideas: A Process for Breakthrough Innovation

June 14, 2023 • 5 min read.

In this Nano Tool for Leaders, Wharton professor Ethan Mollick outlines seven steps teams can take to unlock new ideas.

Nano Tools for Leaders®   —  a collaboration between  Wharton Executive Education  and  Wharton’s Center for Leadership and Change Management  — are fast, effective tools that you can learn and start using in less than 15 minutes, with the potential to significantly impact your success.

Contributor: Ethan Mollick , associate professor of management at Wharton and director of Wharton Interactive . This Nano Tool is based on The Breakthrough Game created by Mollick and game designer Justin Gary.

Adopt a seven-step approach for generating new ideas and solving intractable problems.

Most organizations, when asked to come up with an innovative concept or solution, rely on outdated, unreliable methods like unconstrained brainstorming — if they have any formal approach at all. It’s more common to pass the request along to the person or people who have come up with creative ideas in the past and hope for the best. But decades of research now reveal that innovation is a process involving many important steps, and following them allows individuals and teams to apply the science of innovation quickly and reliably.

Action Steps

1. frame the question.

To decide on the question you want to answer, start by focusing on real results rather than on particular ways of doing things. Asking, “How can we make ordering cheeseburgers faster at our restaurant?” for example, isn’t as effective as focusing on your real end goal: “How might we decrease customer frustration at our restaurant?” Starting questions with “How can we” or “How might we,” helps make them more open-ended, which is in line with advice from innovation expert Hal Gregersen, who says “Questions are most productive when they are open versus closed, short versus long, and simple versus complex.”

2. Build the Team

Teams with diverse backgrounds and experiences, and those with a proven innovative idea track record, are more likely to generate new concepts.

3. Generate Lots of Raw Ideas — Solo

Brainstorming in a group tends to suppress some ideas and surface others that conform to the group. Research shows that you need at least about 100 raw ideas to have the best opportunity for something really innovative.

4. Add Constraints

Studies show that when you have limitations of the ideas you can pursue, you are often, paradoxically, more creative. But don’t just place constraints on the types of ideas you’re looking for — tight time constraints also help by limiting your ability to be overly reflective and judgmental about your own ideas, further increasing creativity.

5. Recombine and Reframe

Come together as a group to discuss the ideas you generated individually. Since innovation often involves recombination, combining old ideas in new ways, aim for big leaps rather than little refinements based on new ideas or novel combinations. Leverage research on brainstorming by prohibiting criticism (ask questions instead of making judgments); encouraging wild ideas (they can be refined later); and improving and combining others’ ideas, not just your own.

Instead of relying on a simple up-or-down process, use categories such as “Biggest Impact,” “Easy Win,” and “I Love This” to winnow down ideas. Be willing to criticize and reject ideas at this stage.

Without follow-up, even the most intriguing innovations die. As you choose which ideas to take further, consider these proven techniques: create a prototype and share it to get feedback; interview potential users/stakeholders; and build consensus by working as a group to uncover potential organizational hurdles and shepherd the idea past them.

How Leaders Use It

Developed in the 1930s by its founder, Toyota’s “Five Whys” technique is a proven brainstorming constraint. The company brings together the people who have direct experience of the process or problem in question, who then define the issue and proceed to ask “Why?” five times. Speed is important as ideas are shared — another constraint that prevents judgment and allows for greater creativity.

Airbnb co-founder Joe Gebbia said in early stages the company was stuck in a “trough of sorrow,” unable to grow. He attributed the stagnation to the belief that everything they did had to be scalable and involve tech. It was only when they started generating ideas to solve the growth problem that they identified poor photo quality of the properties on their website, and allowed themselves to address it by experimenting with non-scalable, non-tech changes. They decided to prototype an idea by having professional photos taken. A week after the new pictures were posted, weekly revenue doubled. This process of framing the question and generating ideas well outside the constraints they previously placed on their ideation efforts, leading to prototyping, “changed the trajectory of the business,” according to Gebbia.

Knowledge in Action: Related Executive Education Programs

• Mastering Innovation: Strategy, Process, and Tools
• Business Model Innovation in the Digital Age
• High-Potential Leaders: Accelerating Your Impact
• The Neuroscience of Business: Innovations in Leadership and Strategic Decisions

Additional Resources

Access all Wharton Executive Education Nano Tools

Download this Nano Tool as a PDF

More From Knowledge at Wharton

The Hidden Business Risks of Humanizing AI

To Survive Sustained Change, Start Rehearsing

How AI Analytics Spurs Innovation at Newly Public Firms

Looking for more insights.

Sign up to stay informed about our latest article releases.

• Niamh Conneely

Marketing Executive

Read Time: 15 minutes

Idea generation is a fundamental process that transcends industries and fields. It’s the driving force behind innovation and progress.

Harvard Business Review  contacted research from data of 3.5 million employees which concluded that higher ideation rates are correlated with growth and net income, most likely because companies with an innovation culture not only generate more unique solutions to market challenges but also encourage a proactive approach to addressing future industry shifts and customer needs.  McKinsey & Company  second this as they state companies that focus on innovation perform much better, generating over 2.4 times profit as other businesses

In this blog, you will be provided with a comprehensive understanding of idea generation, drawing insights from various sources. Whether you’re a student, a professional, an entrepreneur, or simply someone eager to boost their creative thinking, this guide is tailored to help you become a proficient idea generator. We’ll delve into practical techniques, strategies, and tools to empower your creative potential, enabling you to generate fresh ideas that can make a real impact.

• Defining Idea Generation

Idea generation is the vital initial step towards advancement in the fields of creativity and innovation. It’s a skill that can be used in any sector or industry and crosses all barriers. The process of conceiving new ideas, concepts, or possibilities is known as idea generation. The great thing about idea generation is that it is not an exclusive skill, it can be learnt by anybody. It involves going against conventional thinking and embracing the unique and more unexplored ideas.

• The relationship between idea generation and innovation

Innovation and idea generation go hand in hand, supporting each other in the creative journey. When you generate fresh ideas and nurture them, you’re laying the groundwork for innovation.These fresh ideas are the key ingredients of innovation, each holding the potential to transform industries and solve complex challenges. As these ideas evolve, they become the driving force behind groundbreaking innovations—real-world  solutions  that make a difference.

In the upcoming sections of this blog, we’ll explore practical techniques and strategies to help you unlock your idea generation potential.

• Powering your idea generation with Fliplet

Fliplet’s Idea Management solution  is a comprehensive platform designed to streamline idea generation and innovation management processes. It facilitates  collaboration among team  members, enabling them to propose, discuss, and prioritize ideas seamlessly. Fliplet’s solution not only centralizes the entire innovation cycle but also empowers organizations to capture and harness the collective intelligence of their workforce. It acts as a catalyst for innovation, ensuring that promising ideas are nurtured, developed, and transformed into impactful solutions.

• The Significance of Idea Generation for Businesses and Entrepreneurs

Idea generation holds a paramount role in the world of business and entrepreneurship. It acts as a compass guiding organizations and individuals toward success. Here, we’ll explore the vital aspects of idea generation for businesses and entrepreneurs.

• Identifying New Opportunities for Growth and Competitiveness

Idea generation is the driving force behind spotting new opportunities. It’s the process that allows businesses to explore uncharted territories, discover untapped markets, and develop innovative products or services. By continuously generating fresh ideas, businesses can stay ahead in the competitive landscape, ensuring long-term growth and relevance.

• Refinement and Evolution of Existing Practices

Idea generation is not just about reinventing the wheel; it’s also about making the wheel smoother. It enables businesses to refine and evolve their existing practices. By adopting a culture of continuous improvement, organizations can identify areas where they can enhance efficiency, reduce costs, and deliver better value to their customers.

• Importance of Involving the Right Stakeholders for Effective Ideation

Successful idea generation involves collaboration and input from various stakeholders. It’s essential to gather insights from employees, customers, and industry experts who bring diverse perspectives to the table. Engaging the right stakeholders ensures that the ideas generated are well-informed and aligned with the organization’s goals and values. It’s a collaborative effort that maximizes the chances of turning ideas into impactful solutions.

• Techniques and Strategies for Idea Generation

There are a range of techniques and strategies to refresh your idea generation process, paving the way for innovative solutions and fresh perspectives. Some of them include:

• Brain-based Techniques
• Mind Mapping

Mind mapping is similar to creating a visual roadmap of your thoughts. It allows you to organize ideas, concepts, and connections. Its benefits lie in its ability to unravel complexity, aiding in problem-solving and idea generation. Use mind maps when tackling intricate challenges or when brainstorming for creative solutions.

Figma  is a tool you can use to help put your ideas into a mind map.

• First Principles Design

At its core, first principles design breaks problems down to their fundamental elements, stripping away assumptions and preconceived notions. By doing so, it encourages innovative thinking by rebuilding from the ground up. This technique is particularly useful when you need to reframe a problem or seek unconventional solutions.

• Blue Sky Thinking/Brainstorming

Blue-sky thinking, or brainstorming, is the art of generating ideas without constraints. It thrives on the principle of quantity leading to quality. By adopting an open and judgment-free environment, brainstorming allows diverse ideas to emerge, helping you explore a multitude of possibilities and uncover unexpected solutions.

• Collaborative Techniques
• Collaborative Innovation

Collaborative innovation harnesses the collective  intelligence of a team. By bringing diverse minds together, you can draw from a pool of perspectives and experiences, increasing the likelihood of groundbreaking ideas. It’s ideal for addressing complex challenges that benefit from varied viewpoints.

Fliplet’s Innovation Management solution  serves as an excellent collaborative tool by seamlessly bringing together diverse minds, allowing teams to pool their perspectives and experiences, ultimately enhancing the chances of generating new ideas.

• “Yes, and…” (Improvisation Technique)

Inspired by improv comedy, this technique encourages building upon ideas rather than negating them. By accepting and expanding on each other’s suggestions, teams can create innovative solutions. It promotes a culture of creativity and collaboration, making it suitable for idea generation sessions that thrive on synergy.

• Role Playing/Role-Storming

Stepping into different roles allows individuals to explore problems from various perspectives. It’s like viewing an issue through different lenses, which can be particularly valuable when brainstorming solutions that require empathy and understanding of diverse stakeholder viewpoints.

• Analytical Techniques
• The 5W’s and the H

The “Who, What, When, Where, Why, and How” questions help dissect problems thoroughly. This structured approach ensures that you explore all aspects of an issue, providing a well-rounded foundation for idea generation.

• SWOT Analysis

SWOT analysis evaluates Strengths, Weaknesses, Opportunities, and Threats. It’s an effective tool for assessing a situation’s internal and external factors, making it a valuable starting point for generating ideas aimed at capitalizing on strengths and seizing opportunities while mitigating weaknesses and threats.

Download this  free SWOT analysis template .

• Innovative Techniques
• SCAMPER Technique

SCAMPER prompts you to Substitute, Combine, Adapt, Modify, Put to another use, Eliminate, or Reverse aspects of a problem or idea. This versatile technique encourages innovative thinking by challenging you to explore multiple dimensions and possibilities.

Here is the SCRAMPER technique explained

• Opposite Thinking

Opposite thinking encourages considering the exact opposite of conventional solutions. It challenges the status quo and prompts you to examine problems from a completely different angle, often leading to creative and unconventional ideas.

• Brainstorm Cards

Brainstorm cards provide prompts or random words to kickstart idea generation. They serve as creative triggers, inspiring fresh angles of approach and encouraging out-of-the-box thinking.

• Analogy Thinking

Drawing parallels with unrelated concepts can lead to unique insights and innovative ideas. Analogy thinking encourages cross-disciplinary idea generation by connecting seemingly disparate ideas.

Synectic is a technique that combines unrelated elements to create something new. It invites you to explore the synergy between diverse concepts and ideas, often resulting in novel solutions.

• Tech and Digital Aids
• Social Listening

Social listening involves monitoring social media and online conversations to gain insights into customer preferences, trends, and pain points. It provides a real-time window into the thoughts and opinions of your audience, sparking innovative ideas for products, services, or  marketing strategies .

• Idea Management Tools

Idea management tools are digital platforms designed to  streamline the ideation process . These platforms enable teams to submit, collaborate on, and evaluate ideas systematically. They promote organization and transparency in idea generation, ensuring that promising concepts are nurtured and developed.

• Digital Platforms for Idea Collaboration

These platforms facilitate remote idea generation and collaboration, allowing teams to work together seamlessly regardless of their physical locations. They promote inclusivity and diverse perspectives, enhancing the quality and quantity of generated ideas.

Discover how a  Fliplet  client launched an idea collection app, achieving 170% above their target:

• Advanced Techniques
• Complex Opportunity Recognition

Identifying intricate opportunities within challenges requires a keen eye for detail and a holistic view of the situation. This advanced technique involves deeply analyzing complex problems to uncover hidden possibilities.

• Product Trend Approach

Analyzing market trends and consumer behavior can lead to innovative product ideas that align with current demands. By staying attuned to shifts in consumer preferences, you can identify opportunities for product innovation.

• Innovation Mapping

Innovation mapping involves visualizing your innovation journey from idea to implementation. It helps you plan and strategize the path your idea will take, ensuring a smooth transition from concept to reality.

Here’s a video explaining innovation mapping:

• Business Model Innovation

What makes Amazon, Spotify and Tinder so successful in such complex industries? The answer is Business Model Innovation. Take a look at this video to find out more:

Reimagining your business model can open doors to new markets and revenue streams. It involves exploring alternative ways to deliver value to customers and generate revenue, often resulting in innovative business strategies.

• Storyboarding

Visualizing ideas through storytelling allows for a more comprehensive exploration of concepts. It helps teams understand the user experience, functionality, and potential challenges associated with an idea.

• Forced Relationships

Forced relationships involve connecting seemingly unrelated concepts or ideas to create new insights. It encourages thinking outside the box and exploring unconventional connections.

• Six Thinking Hats and other Decision-Making Models

These models provide structured frameworks for thinking about problems and generating ideas. They guide individuals and teams through different modes of thinking, ensuring comprehensive exploration of ideas from various angles.

• Techniques Beyond the Norm
• Reverse Brainstorming/Thinking

Reverse brainstorming involves identifying ways to worsen a problem or create obstacles. By considering what not to do, you can uncover innovative solutions to overcome challenges.

• Brainwriting

Brainwriting is a written form of brainstorming where individuals silently contribute ideas on paper. It promotes equal participation and can be less intimidating than traditional brainstorming sessions.

• Brain netting

Similar to brainwriting, brain netting involves collecting ideas from multiple contributors but in a structured and collaborative manner. It encourages free thinking and diverse contributions.

• Group Sketching

Group sketching involves visually representing ideas through drawings or diagrams. It’s particularly effective for teams with visual thinkers who can convey concepts more effectively through visuals.

• Trigger storming

Trigger storming uses cues or triggers, such as images or keywords, to stimulate idea generation. It’s a creative technique that encourages spontaneous and imaginative thinking.

• “What if” Scenarios

“What if” scenarios involve exploring hypothetical situations or scenarios. By asking “what if” questions, you can uncover unconventional ideas and innovative solutions.

A zero draft is an initial, unstructured version of an idea or concept. It encourages free-flowing creativity without the constraints of perfection, allowing for raw and unfiltered ideation.

• Selecting the Right Technique for your Team and Organization

Idea generation isn’t a one-size-fits-all endeavor. It’s a dynamic process influenced by various factors, including team size and specific goals. In this section, we’ll explore the art of choosing the right idea generation technique to match your team’s composition and objectives.

• Selecting the Right Idea Generation Technique: Team Size vs. Goal

The path to effective idea generation begins with understanding your team’s dynamics and the goals you aim to achieve. The selection of the appropriate technique hinges on two key considerations: team size and the nature of your goal.

• Tools that Can Help in Idea Generation and Innovation Management

Leveraging the right tools can significantly enhance the efficiency and effectiveness of your idea generation and innovation management efforts. Here are three essential tools, each playing a vital role in nurturing creativity and driving innovation:

When it comes to idea generation, having the right tools can significantly boost your creative potential. That’s where Fliplet can help.  Fliplet offers an Innovation Management solution  that caters to both individuals and teams. Fliplet can be a valuable platform in your creative journey. Fliplet provides a structured platform for brainstorming, collaborating, and nurturing your ideas.

Fliplet’s  innovation solution  features can help spark ideas. Some key features include:

• Easily ask employees for solutions to urgent problems in your organization

Fliplet’s  innovation management software  allows you to swiftly address pressing concerns within your organization. By creating specific challenges, you stimulate solution-oriented thinking among your team. This feature promotes proactive problem-solving, nurturing a culture of innovation.

• Prioritize the innovative ideas with the biggest business impact

You can tap into the full potential of your workforce. It empowers your team to engage with each other’s ideas, allowing them to like, comment, and prioritize the most impactful ones. This collaborative approach ensures that your organization’s future is shaped based on collective employee intelligence.

• Give your employees a place to propose ideas for solving your organizational challenges

Fliplet provides a dedicated platform for sparking creativity and open innovation. By offering a space for sharing and discussing ideas, it encourages employees to contribute innovative solutions. This feature creates a culture where every team member feels empowered to propose and explore fresh ideas.

• Manage your entire innovation cycle in one view

Oversee your entire innovation cycle, from ideation to implementation, all within a single app. By embracing this streamlined approach, you enhance efficiency and effectiveness in your innovation endeavors. Fliplet ensures that your ideas seamlessly progress from concept to reality, making innovation a smoother journey.

With Fliplet, you can streamline your idea generation process, organize your thoughts effectively, and turn those concepts into  actionable plans . Whether you’re working solo or within a team, Fliplet can be a catalyst for harnessing your creative potential. So, stay tuned as we uncover how Fliplet’s innovation solution can elevate your idea generation game. See how Fliplet can help you drive impactful innovation,  try Fliplet for free .

Another software you may want to use is  Figma . Figma is a versatile design and collaboration tool that can be effectively used for mind mapping. Its  collaborative features allow teams  to visually brainstorm ideas and create mind maps in real-time. Figma’s flexibility and cloud-based platform make it an ideal choice for remote teams to collaborate on ideation and organize their thoughts efficiently.

Thirdly,  IdeaScale  is a dedicated idea management software that provides organizations with a platform to collect, evaluate, and implement ideas systematically.

These tools are instrumental in innovation management by providing a structured framework for idea collection, evaluation, and implementation. They encourage collaboration, transparency, and the efficient  management of innovative projects , ultimately driving the growth and success of organizations.

• Idea Generation in Specialized Fields

Idea generation isn’t limited to any particular industry or field; it’s a universal process that can be applied across various domains. Here, we explore how idea generation plays a crucial role in two specialized fields: entrepreneurship and product development.

• Entrepreneurship

Idea generation is the foundation upon which innovative startups are built. Entrepreneurs constantly seek unique solutions to existing problems or identify untapped opportunities in the market. Idea generation in entrepreneurship involves market research, identifying pain points, and envisioning novel products or services that can address these needs.

Entrepreneurs often rely on brainstorming sessions,  customer feedback , and industry trends to generate innovative ideas that can disrupt traditional markets or create entirely new ones. The ability to generate and refine ideas is a hallmark of successful entrepreneurship.

• Product Development

Idea generation is a fundamental aspect of product development. Companies engaged in product development must continuously innovate to stay competitive and meet the evolving needs of their customers. Idea generation in this context often involves cross-functional teams working together to ideate and conceptualize new products or enhance existing ones.

Techniques such as design thinking, rapid prototyping, and user-centered design play a pivotal role in generating ideas that can lead to the creation of successful products.  Product development teams  focus on translating customer insights, market trends, and technological advancements into actionable ideas that result in tangible products with a competitive edge.

• Navigating Challenges in Idea Generation

Idea generation, while essential for innovation, comes with its set of challenges. Understanding and overcoming these hurdles is critical for a successful ideation process. In this section, we’ll explore some common pitfalls and misconceptions and emphasize the role of feedback and iteration in idea generation.

• Common Pitfalls and Misconceptions
• Overemphasis on Perfection:  One common pitfall is striving for perfect ideas from the outset. This can lead to a paralysis of analysis and hinder creativity. In reality, ideas often start as rough concepts and evolve through iterations.
• Lack of Diverse Perspectives:  Restricting ideation to a narrow group can limit the variety of ideas generated. Diverse perspectives, backgrounds, and experiences are essential for innovative thinking.
• Ignoring Feedback:  Failing to seek and incorporate feedback from peers, stakeholders, or customers can result in missed opportunities for improvement and refinement.
• Fear of Failure:  The fear of failure can stifle creativity. Embracing the idea that not all ideas need to succeed can free individuals and teams to explore more daring and innovative concepts.
• Rigid Thinking:  Sticking to preconceived ideas or established processes can hinder creativity. Being open to new approaches and possibilities is crucial for idea generation.
• The Role of Feedback and Iteration
• Enhanced Idea Quality:  Feedback provides valuable insights and perspectives that can help refine and enhance the quality of ideas.
• Validation:  It ensures that ideas align with organizational goals, customer needs, and market trends, reducing the risk of pursuing unviable concepts.
• Continuous Improvement:  Iteration allows for continuous improvement of ideas. Concepts often start as rough drafts and evolve through cycles of refinement and testing.
• Risk Mitigation:  Iterative processes help identify and mitigate potential risks and challenges associated with implementing ideas.
• Alignment with Objectives:  Feedback and iteration ensure that ideas remain aligned with organizational objectives and strategies.

Challenges in the idea generation process can be effectively addressed with Fliplet’s innovation solution, a comprehensive tool designed for the nuances of ideation. Fliplet’s platform facilitates collaboration, captures feedback, and streamlines the iteration process, ensuring that valuable ideas are refined and transformed into impactful solutions.

• Measuring and Managing Your Ideas

Effectively measuring and managing ideas are essential aspects of the idea generation and innovation process. In this section, we’ll explore key performance indicators (KPIs) and metrics for evaluating idea quality and viability, as well as provide tips for successfully managing and refining ideas.

• KPIs and Metrics for Evaluating Idea Quality and Viability
• Idea Impact Score:  Assess the potential impact of an idea on your organization. Consider factors like potential revenue, cost savings, or improved customer satisfaction.
• Feasibility Score:  Evaluate the feasibility of implementing the idea. Assess whether the necessary resources, skills, and technology are available.
• Customer Feedback and Validation:  Measure the level of positive feedback and validation from customers or target users. Customer surveys, interviews, and user testing can provide valuable insights.
• Time-to-Value:  Evaluate how quickly an idea can be implemented and start delivering value to the organization.
• ROI Projection:  Estimate the return on investment (ROI) the idea is expected to generate over a specific period.
• Alignment with Strategic Objectives:  Assess how well the idea aligns with the organization’s strategic objectives and long-term goals.
• Cost-Benefit Analysis:  Compare the potential benefits of the idea against the estimated implementation costs.
• Tips for Successfully Managing and Refining Ideas
• Establish Clear Idea Ownership:  Assign responsibility for each idea to a specific team member or stakeholder to ensure accountability.
• Regular Idea Reviews:  Conduct regular reviews and discussions to track the progress of ideas and identify areas for improvement.
• Prioritization:  Prioritize ideas based on their impact, feasibility, and alignment with organizational goals.
• Iterate and Refine:  Encourage continuous refinement and iteration of ideas based on feedback and changing circumstances.
• Create Cross-Functional Teams:  Form cross-functional teams to work on ideas, leveraging diverse skills and expertise.
• Document and Archive Ideas:  Maintain a centralized repository of ideas, even if they are not immediately pursued, as they may become valuable in the future.
• Communication and Transparency:  Ensure clear communication and transparency throughout the idea management process to keep all stakeholders informed.
• Celebrate Successes:  Recognize and celebrate the successful implementation of ideas to motivate teams and enhance a culture of innovation.

Effectively measuring, managing, and refining ideas using these KPIs and tips can lead to the successful execution of impactful ideas, driving innovation and growth within your organization.

• Building an Effective Idea Generation Team

Creating an effective idea generation team is pivotal for successful innovation within any organization. To achieve this, start by assembling a team with diverse skills and backgrounds to bring varied perspectives and creativity to the table. Encourage collaboration among team members from different departments to approach challenges comprehensively, creating a culture of cross-functional cooperation.

Designate passionate individuals as innovation champions to inspire and motivate the team, driving the ideation process forward. Define clear roles and responsibilities within the team to ensure a structured approach and accountability. Invest in training and development programs to equip team members with idea generation tools and techniques, generating continuous  learning  and industry awareness. Implement incentive and recognition programs to reward innovative contributions and boost morale.

Provide access to idea management tools like Fliplet to facilitate collaboration, capture ideas, and  track progress  efficiently. Schedule frequent ideation sessions to encourage brainstorming and idea sharing, cultivating a culture of open and constructive feedback among team members. Enable hands-on experience by allowing team members to prototype and test ideas, and encourage exploration of ideas from other industries or domains to inspire innovation. Ensure leadership provides the necessary support and resources for the idea generation team, as leadership buy-in is crucial for the success of innovation initiatives. By following these steps, organizations can build a dynamic idea generation team poised to drive impactful innovation.

• Why Idea Generation is so Important

In conclusion, adopting a culture of continuous learning and innovation is essential for organizations looking to stay competitive and drive growth. It’s crucial to encourage your teams to explore new ideas and experiment with various techniques to unlock their full creative potential.

Additionally, consider leveraging tools like  Fliplet’s Innovation Management solution , which can streamline the idea generation process, facilitate collaboration, and help manage and refine ideas effectively. Embrace innovation as an ongoing journey, and you’ll be well-positioned to tackle future challenges with fresh, innovative solutions.

Niamh Conneely is the Content Writer and Marketing Executive at Fliplet – an app building platform that enables anyone to easily create engaging and interactive mobile and web apps.

Niamh has experience in content creation and writing and has achieved two level 6 awards for The Chartered Institute of Marketing. She has developed skills to help organizations deal with the challenges of a fast-moving marketplace. Connect with Niamh on Linkedin.

• What is idea generation?

Idea generation is the process of coming up with new and creative concepts or solutions. It’s like brainstorming sessions where people think of innovative ways to solve problems or create something new.

• Which idea generation method involves keeping ideas in sight and well-organized?

The method called “Mind Mapping” helps keep ideas in sight and organized. It’s like drawing a tree with branches; the main idea is the trunk, and related thoughts or sub-ideas branch out from it. This visual representation helps people see connections and organize their thoughts clearly.

• What takes place during the idea generation stage of the new-product development process?

During this stage, people brainstorm and come up with various concepts for a new product. It’s like planting many seeds and seeing which ones have the potential to grow. The aim is to gather as many ideas as possible and then pick the best ones to develop further.

• FirstHeading

Creating your app has never been easier

See the latest insights from fliplet, ultimate guide: how to scale mobile enterprise app development without coding or it resources, top 10 mobile enterprise application platforms, the best low code/no code development platforms, creating your app has never been easier..

Fliplet compiled an ultimate guide on how to scale mobile enterprise app development without coding or IT resources

Fliplet compiled a list of the top 10 mobile enterprise application platforms. Read more within.

Fliplet compiled a list of the best Low Code/No Code Development Platforms. Find out more within.

Wow someone’s app is really popular!

Let’s find you a plan that keeps up with your ambition. ⭐

43 brainstorming techniques and games for creating new ideas

Finding new and innovative ideas is a vital part of the growth and success of any team or organization . While brainstorming techniques are rightly perceived as creative and exciting , it’s important to find a framework and idea-generation process that empowers your group to generate meaningful results. 

Innovation is important for many businesses, but what brainstorming activities might you use to help make true innovation a reality? Find out in this collection of effective brainstorming techniques!

Design your next session with SessionLab

Join the 150,000+ facilitators 
using SessionLab.

Recommended Articles

A step-by-step guide to planning a workshop, 54 great online tools for workshops and meetings, how to create an unforgettable training session in 8 simple steps.

• 18 Free Facilitation Resources We Think You’ll Love

In this post, we’ll explore a host of effective brainstorming techniques in categories such as creative exercises and visual idea generation games, all of which can be used to help your group brainstorm be more effective and gratifying for all involved.

We’ll also explore talk about the benefits of group ideation and share some examples of brainstorming sessions that utilize these methods. Let’s get started! 

What are brainstorming techniques?

Brainstorming is a process of enabling people to think freely and creatively when trying to come up with ideas, solutions, or sharing knowledge.

Brainstorming techniques are proven activities and frameworks for coming up with lots of ideas quickly. They’ll often include steps to shift perspective, facilitate team collaboration and refine initial ideas into something even better.

Some examples of brainstorming activities include classic mind mapping and brain writing where you quickly try to generate as many ideas as possible.

Teams often use these techniques to generate creative ideas to tough problems and to explore possible solutions . They can also be used as part of more in-depth brainstorming workshops where team members then refine and choose ideas to put into action.

Some of the core concepts of brainstorming include: reserving judgment, go for quality over quantity, listen to all ideas, and think outside of the box in the pursuit of radical new ideas and creative solutions. Bring these concepts and a proven technique to your session and you’re already on the route to success!

If you’re finding your team with a problem they don’t know how to solve, a technique that encourages creative thinking might be just the ticket! Use these activities as part of a complete workshop process to refine those ideas into something actionable.

In SessionLab, it’s quick an easy to create an effective agenda for a brainstorming workshop in minutes. Drag and drop blocks in the session planner to create your structure. Add timing for each item to ensure you stay on time. Color-code your blocks to get an instant sense of your learning flow.

Core group brainstorming techniques

Brainstorming has been around as long as individuals and teams have tried to find creative and innovative solutions, or come up with new ideas or products. Whether a group is ideating on how to solve an organizational problem or generate ideas for new features or initiatives, getting people together to quickly ideate and come up with something new is time well spent.

In this section, we’ll first explore some of the core techniques that have been used time and again to create meaningful results and great ideas.

Popcorn Brainstorming

One of the classic brainstorming techniques, chances are you’ve done a popcorn brainstorm already. It’s been used by everyone from school teachers to CEOs to generate ideas and create energy around new initiatives – much like the popping of corn in a microwave!

Start by posing a question or problem statement and invite participants to take a minutes silence to think on it. Once the minute is up, start a timer and invite everyone to contribute ideas out-loud and build on each other’s ideas too. Have a single person take notes and encourage quality over quantity: no evaluation, no criticism or discussion yet – just rapid ideation!

Brainstorming – Popcorn and Round Robin   #idea generation   #brainstorming   Simple, classic brainstorming with two variants. Popcorn – where participants speak out-loud and Round Robin – where participants work in silence and pass their ideas to the next person in turn. 

Round-Robin Brainstorming

A tried and tested idea generation technique, Round-Robin Brainstorming provides a little more structure and ensures everyone in a group can contribute to a brainstorm by ensuring the discussion isn’t dominated by the loudest voices. 

In this group method, seat everyone in a circle and hand them an index card. In silence, everyone writes an idea on their index card before passing it to the person to their left. Each participant then writes an idea based on their neighbour’s card and passes that along.

The result is a more relaxed session that encourages a combination of idea development and co-creation while ensuring everyone is heard. Perfect for teams with big personalities!

SWOT analysis

A SWOT analysis (strengths, weaknesses, opportunities and threats) is a tried and tested technique that teams often using when planning new initiatives or solving problems. It also happens to be a great tool for generating new ideas while also taking into account potential problems and opportunities.

The act of brainstorming around your weaknesses or threats can result in innovative solutions and ideas you might not have otherwise come up with. Try using each point of the process as a jumping off point for ideation or explore a topic from each of the different angles for best results.

SWOT Analysis   #project planning   #strategic planning   #environmental analysis   #planning   #issue analysis   #online   #remote-friendly   A SWOT Analysis is used in project planning, strategic planning and other processes where agreement is needed about the current situation of a project, team, department or organization. It stands for Strengths, Weaknesses, Opportunities and Threats.

Brainwriting

No single person is as smart or as creative as a group. With the brain writing technique, tap into the ideas and approaches of multiple collaborators and co-create effectively.

Start by writing down the topic or area for which you’ll be generating ideas. Have all team members silently write down an idea related to the topic on a card and then, pass that idea to the person to their right. The receiving player reads the card and then adds an idea inspired by the original OR enhances the original idea before passing the card along.

By asking participants to grow and improve on one another’s ideas, Brain writing helps a group ideate effectively and come up with better ideas. You can even bring this to an online brainstorming session by using an online whiteboard and have participants pass post-its to the working spaces of their partners.

Brainwriting   #gamestorming   #idea generation   Some of the best ideas are compilations from multiple contributors. Brainwriting is a simple way to generate ideas, share them, and subsequently build on them within a group. Access to multiple hands, eyes, and minds can yield the most interesting results.

Question storming

Sometimes, shifting perspective and starting from a different angle can generate the best ideas. Q-Storming, or question storming invites participants to brainstorm questions, rather than solutions.

After rounds of gathering qualifying data and assumptions, ask your group to think of all those questions that they still have which might help the team think the matter through. This approach can be really useful at finding ideas your team might not have considered and ensuring that what you come up with is truly going to solve the problem at hand.

Walking Brainstorm

Brainstorming methods come in many forms – you might have a quick-fire session that encourages excitement and verbal exchanges. Alternatively, you might find your group will create better ideas by working together in a more relaxed, introvert-friendly manner.

Walking Brainstorm is designed to help large groups work on idea generation dynamically but without creating scenarios where only the loudest participants are heard. 

Create a space where different topics or questions are spread on posters/post-its around a room or virtual space. Silently and individually, each participant is encouraged to walk around and visit each question/topic in turn and add ideas to each. By moving around and working individually, this method helps create a more reflective, dynamic ideation session and can also help ensure group-think doesn’t set in! 

Walking Brainstorm   #brainstorming   #idea generation   #remote-friendly   This introvert-friendly brainstorming technique helps groups of any size to generate and build on each other’s ideas in a silent but dynamic setting. As the participants keep moving, the exercise is ideal to kick-off a full day workshop or re-energize the group after lunch.

For those who prefer a more organized approach to idea generation, mind mapping is a great activity for creating ideas quickly and effectively. 

Begin by writing the key topic in the center of a piece of paper or in an online whiteboard. Invite participants to brainstorm related topics and ideas by adding branches to the central idea and create new nodes or elements. As a facilitator, you’ll want to group ideas by color and also amend the thickness of the branches to show the strength of various ideas and concepts. 

When you’re done mind mapping, the result will be a diagram that visually represents your ideas and makes it clear how the various parts interrelate – a great resource for idea development or for future sessions!

Mind map   #idea generation   #concepts   #create   #issue analysis   #design   A mind map is a diagram used to represent a number of ideas or things. Mind maps are methods for analyzing information and relationships.

Brain Netting

The concept of brain netting is to not only take your brainstorming online, but to use online tools and virtual spaces to make the session a truly engaging experience. 

The key is to use an online tool that the group is familiar with, can co-create in easily and which works both synchronously or asynchronously. Using an online whiteboard or shared document for brain netting means participants can contribute in both a live online workshop as well as in their own time. This is a unique benefit of online brainstorming, and it’s one we’d recommend taking advantage of with your team!

We’d especially recommend using an online tool that supports easy commenting, images, videos and links – encourage your group to use whatever assets best communicate their ideas!

Six Thinking Hats

Exploring a problem or idea from multiple perspectives is a great way to generate new ideas and inform your brainstorming process. In this brainstorming activity, start by explaining the six different hats and that at various points, each person will wear the different hats to explore your chosen topic. For example, the green hat is for creative thinking while the white hat is all about information and facts.

Invite the group to start with the blue hat, which is to control the process and then move between hats to explore, define, ideate, identify risk and gather information around a topic in a sequence. By asking the group to all wear the same hat at the same time, you can ensure your brainstorm moves forward while also ensure all perspectives are explored.

The Six Thinking Hats   #creative thinking   #meeting facilitation   #problem solving   #issue resolution   #idea generation   #conflict resolution   The Six Thinking Hats are used by individuals and groups to separate out conflicting styles of thinking. They enable and encourage a group of people to think constructively together in exploring and implementing change, rather than using argument to fight over who is right and who is wrong.

Rapid writing

Different teams and workshops need different approaches to generating fresh ideas. While a carefully structured approach can be effective, using quick-fire brainstorming techniques like Rapid Writing can help create a sense of energy, urgency, and get heaps of ideas out quickly.

For this method of brainstorming, start by setting a timer and encouraging your participants to get as many ideas out as possible within that time limit. Remember that at this stage in the idea generation process, there is no such thing as a bad idea and by quickly ideating without being critical, your group can be creative without prematurely shutting down possible ideas. Be sure to collect all the ideas and share them without judgment at the end, whether you’re brainstorming online or in person!

Lotus Blossom

Some of the most effective techniques are those that encourage free-thinking and rapid ideation while also having some rules that can keep things structured. Lotus Blossom combines these concepts while also creating a great visual representation of your brainstorming activity. 

Lotus Blossom helps facilitate idea generation by working out from a central concept and adding eight additional themes or ideas inspired by the first on sticky notes. Once you have those eight ideas, you then invite participants to take each of those and add another eight and effectively blossom them around the original. By clustering ideas in this way, this ideation method also creates a visual resource you can come back to later and follow the brainstorming process from start to finish.

Lotus blossom   #concepts   #create   #design   #idea generation   The lotus blossom method is a creativity exercise. It is a framework for idea generation, starting from one central theme. Eight conceptual themes grow out from the main theme and each of them are used as central theme to generate 8 more themes. Explore!

Starbursting

Complete freedom without an ideation framework isn’t always the best way to find and develop ideas. Structured techniques like Starbursting can help guide a team through more effective idea generation and ensure all key elements are considered at an early stage.

To begin, create a six-pointed star on a large piece of paper or online whiteboard. At the tip of each point of the star, write down the words Who, What, Why, Where, When and How. Invite the group to brainstorm ideas and questions related to each of these points in turn.

At this stage, the group only needs to brainstorm questions in each of these sections, leaving answers until later, though creating follow-up questions can also be helpful in effectively ideating on your central concept or problem.

When ideating on solutions to problems, it’s very easy to come to the table with underlying assumptions that can affect the course of the idea generation process. You can avoid this potential pitfall by using The 5 Whys to go further and deeper in a very simple, group-friendly manner.

Kick-off by working as a group to create a problem statement that you’ll work on solving. Once you have a concise statement, ask the group why you have this problem and discuss the answer. After working together to form a cohesive answer, ask the group why you have the problem again. By repeating the process, you and your team can dig deeper and find the root cause of the issue and move past the first, most obvious ideas.

The 5 Whys   #hyperisland   #innovation   This simple and powerful method is useful for getting to the core of a problem or challenge. As the title suggests, the group defines a problems, then asks the question “why” five times, often using the resulting explanation as a starting point for creative problem solving.

Creative brainstorming techniques 

All brainstorming is creative. Generating ideas and finding solutions often asks groups and teams to find new ways of looking at things but in this next section, we’ll look at techniques that aim to approach the ideation process from a unique or especially creative starting point.

If you’re finding your typical exercises aren’t yielding results or want to try something new, creative games like those below can create space for innovation. Let’s dig in!

Imagie-ination

Words are often our primary tool when it comes to starting a brainstorm or kicking off an idea generation workshop. While these kinds of techniques are tried and tested, it can also be useful to try something different that can unlock your team’s creativity. 

With Imagie-ination, you’ll use images to help your group generate ideas that go beyond the norm. First, collect an assortment of images for your brainstorm and write down a simple description of the topic you want to generate new ideas around. Have each participant select an image and then come up with as many ideas as they can for how the image relates to the topic. 

After the first round, you’ll then cluster ideas together and find an image and title to best illustrate those clusters. This kind of clustering and titling can help refine the ideas your team has generated and move them towards action – a great outcome for any brainstorming session!

Imagie-ination   #idea generation   #gamestorming   Images have the ability to spark insights and to create new associations and possible connections. That is why pictures help generate new ideas, which is exactly the point of this exercise.

Bad idea brainstorming

Idea generation is at its best when groups are encouraged to add their ideas without being self critical or overthinking. Often, individuals involved in ideation can put pressure on themselves to offer only great ideas and so don’t contribute everything that comes to their minds.

Use this brainstorming technique to help free your group’s creativity and encourage them to come up with the absolute worst ideas they can in relation to a central topic or problem. Like reverse brainstorming, this brainstorming technique is a great way to find alternative routes to more creative ideas. Just be sure to use a swot analysis to figure out what should make it into reality!

Bad Idea Brainstorm   #brainstorming   #creative thinking   #idea generation   Name all the bad ideas to make room for good ones. Coming up with the perfect solution right off the bat can feel paralyzing. So instead of trying to find the right answer, get unstuck by listing all the wrong ones.

Brainstorm questions instead of solutions

Our first instinct when it comes to problem-solving can often be to jump straight to giving answers and finding solutions. Though this can be effective, when it comes to generating creative ideas, a different tact can be more effective. 

With this reverse brainstorming game, challenge participants to offer questions instead of solutions so they can respond to a central concept creatively and from a new angle. If you’re finding your group can become blocked when generating ideas, it might be that your existing questions or frameworks aren’t sufficient. By taking a new ideation approach, you can unblock your team!

Brainstorm questions instead of solutions   #questions and answers   #brainstorming   When we are given a problem our reflex is to find answers. But it can be difficult to leave the comfort zone and to come up with creative answers. This exercise will encourage to think out of the box.

Stakeholder Round Robin Brainstorm

Bringing together groups of different stakeholders with their own areas of expertise is a great idea whether you’re brainstorming or finding solutions. That said, it’s worth noting that in these kinds of mixed groups, participants will be coming from different places and have different priorities and approaches to idea generation. 

Start by creating a flipchart or whiteboard space for each stakeholder’s perspective and writing this at the top. Give each stakeholder two minutes to brainstorm on the central idea from their perspective and add those to their flipchart before then inviting each participant to move to the next one and brainstorm from this different perspective. By using this round-robin brainstorming exercise, you can help the group understand the perspectives and insights each member brings to the table while also generating fresh ideas as a result!  

Backcasting

A simple change in perspective can have a massive impact on how your team approaches solving a problem. Backcasting is a simple but effective brainstorming exercise where a team is invited to work backwards from an ideal future state in order to come up with concrete actions they can take today.

Start by listing your long term goals in a time frame of 1-20 years. Then work backwards from that state to today, listing every action necessary to achieve that goal state. Collect insights on what difficulties might come up, what steps your team needs to take and what resources you might need in order to brainstorm effectively and find a new way to reach your team’s long term goals!

Backcasting   #define intentions   #create   #design   #action   Backcasting is a method for planning the actions necessary to reach desired future goals. This method is often applied in a workshop format with stakeholders participating. To be used when a future goal (even if it is vague) has been identified.

Walking Questions

A brainstorming technique with a What if learning style, Walking Questions is a great way of encouraging group members to share knowledge, ask questions to personal problems, and explore a topic dynamically. 

Best used at the end of a training session or workshop, each participant writes a question they have on the top of a sheet of paper then hands it to the person to their right. The person receiving the paper then writes any ideas or answers they have underneath and passes it to the next person.

By the end, the original piece of paper will be returned to the owner filled with ideas and answers from the entire group. It’s a great way of generating ideas from a group quickly and efficiently and of utilizing everyone’s expertise in a structured way. Give it a go! 

Walking questions   #what if learning style   #idea generation   #learning   This is a great facilitation technique to answer open questions of trainees with a “What if” learning style. It prevents the facilitator from answering all questions herself. With this method trainees can:  close knowledge gaps find solutions for personal problems imagine themselves using their new knowledge in future and prepare themselves for obstacles

Guided Imagery

Coming up with new ideas doesn’t always have to be boisterous! You can also get the creative juices flowing in a relaxed way by tapping into mindfulness and imagination with this method.

Start by inviting participants to close their eyes and get comfortable. Next, progress through a guided meditation designed to inspire creativity. Afterward, ask your group to reflect on what came up for them in the meditation and use this as the basis for further brainstorming!

Guided Imagery   #idea generation   #creativity   #online facilitation   #reflection   This can be used for idea generation especially when the group is stuck.

Headlines from the future

Starting from the desired outcome and working backward with a reverse brainstorming technique can be a great way to solve a problem. Thinking into the future can also be inspiring in a way that encourages free thinking and big ideas – a great result for any brainstorming workshop.

In this idea generation game, ask your group to imagine it’s twenty years in the future and that your project or organisation has been a huge success. Invite each participant to draft a headline and sketch an image for a New York Times feature of this reality. Encourage big, bold ideas and debrief by discussing any common themes or ideas before moving onto idea development as a team! 

Headlines from the Future   #creative thinking   #design   #idea generation   #creativity   Get inspired today by a world 20 years away. Sometimes it helps to start from the end. This exercise will help you align with your team on an audacious vision for your project – one that you can work backward from.

Brainstorming techniques for problem solving and refining ideas

When you want to go beyond initial brainstorming and generate more refined ideas, the following complex idea generation techniques can effectively guide you through the process.

These activities combine brainstorming with idea evaluation, idea selection, and then going into concept development to help you come up with the best options. Let’s dig in! 

25/10 Crowd Sourcing

Group ideation can be tricky to manage, and not all techniques are up to the task of managing creative input from large groups effectively. 25/10 Crowd Sourcing is a fantastic exercise that not only invites big, bold ideas, but can ensure everyone takes part in generating ideas as a group.

After first inviting participants to write a big, bold idea on an index card, start a timer and invite the group to move around the space and exchange cards without reading. Stop the timer and ask each person to read the idea and give it a score from 1 to 5. Repeat five times so that each idea has a score out of twenty-five and then find and share the top ten ideas with the group.

Group brainstorming techniques with a mix of blind scoring and sharing can be especially useful in avoiding bias and encouraging bold ideas – especially useful when ideating in large groups!

25/10 Crowd Sourcing   #idea generation   #liberating structures     You can help a large crowd generate and sort their bold ideas for action in 30 minutes or less! With 25/10 Crowd Sourcing , you can spread innovations “out and up” as everyone notices the patterns in what emerges. Though it is fun, fast, and casual, it is a serious and valid way to generate an uncensored set of bold ideas and then to tap the wisdom of the whole group to identify the top ten. Surprises are frequent!

3-12-3 Brainstorm

Brainstorming is often associated with fast ideation and energetic idea generation sessions. While many standard techniques can be slowed down and run in different ways, there can be obvious benefits to maintaining energy and proving the value of short working bursts to your ideation group. 

The 3-12-3 Brainstorm technique taps into the power of speed to generate great ideas and can help a team generate, develop, and present ideas in just less than twenty minutes. By combining speed and structure, this ideation method can help a group pressed for time use the session effectively, and we love how much ground can be covered by a group with this exercise!

3-12-3 Brainstorm   #gamestorming   #idea generation   This format for brainstorming compresses the essentials of an ideation session into one short format. The numbers 3-12-3 refer to the amount of time in minutes given to each of three activities: 3 minutes for generating a pool of observations, 12 for combining those observations into rough concepts, and 3 again for presenting the concepts back to a group.

Mash-Up Innovation

Some of the best ideas come from taking existing ideas and putting them together. Brainstorming that takes advantage of what your group already knows and loves can really supercharge the idea generation process and this creative exercise is a perfect example of that.

In Mash-Up innovation, first ask your group to brainstorm around three different topics or areas and add them to a shared space or whiteboard. Next, organize your participants into small groups who will spend the next twelve minutes combining and mashing up as many of the elements as possible to make even better ideas. After a short idea presentation, you can even take your group through an idea development stage to really make the most out of this activity. You’ll be surprised at what comes out!

Mash-Up Innovation   #hyperisland   #innovation   #idea generation   Mash-ups is a collaborative idea generation method in which participants come up with innovative concepts by combining different elements together. In a first step, participants brainstorm around different areas, such as technologies, human needs, and existing services. In a second step, they rapidly combine elements from those areas to create new, fun and innovative concepts. Mash-ups demonstrates how fast and easy it can be to come up with innovative ideas.

Large group brainstorming can be dynamic, exciting, and productive but without structure and strong facilitation, it can also become disorganized and frustrating. 1-2-4-All is a classic idea generation process that not only helps ideas find space to allow idea development but also ensures the entire group can contribute to the session. 

Start by inviting silent self-reflection on a shared challenge or question before then moving to pairs, foursomes, and then entire group ideation. One of the many benefits of this brainstorming technique is that everyone gets a chance to contribute freely and share their ideas in a structured way. In groups where conversation can become dominated by strong personalities or not everyone gets a chance to speak, this method is well worth employing. 

1-2-4-All   #idea generation   #liberating structures   #issue analysis   With this facilitation technique you can immediately include everyone regardless of how large the group is. You can generate better ideas and more of them faster than ever before. You can tap the know-how and imagination that is distributed widely in places not known in advance. Open, generative conversation unfolds. Ideas and solutions are sifted in rapid fashion. Most importantly, participants own the ideas, so follow-up and implementation is simplified. No buy-in strategies needed! Simple and elegant!

One will get you Ten

Ideas most often spring from other ideas, and the sharing and co-creation of ideas during a brainstorm or idea generation workshop is where the magic really happens. This technique uses the sharing of ideas between teams as a central concept and it’s a great way of having participants communicate and learn from one another’s ideas meaningfully. 

After first generating ideas solo and as a team, this ideation game asks team members to pair up with members of another team and share ideas. Each member then returns to their own team and presents two ideas – one of their own, and one from the other team – while the rest of the team guesses whose is whose. It’s interesting to see how participants package and present generated ideas and find ways to improve them organically as part of a rapid ideation.  

One will get you Ten   #thiagi   #idea generation   #team   If I give you a dollar and you give me a dollar, we both end up where we began. But if I give you an idea and you give me an idea, we end up with two ideas each, benefiting from a 100 percent return on our investment. In One Will Get You Ten, we leverage this principle so that you and all other participants receive a 1000 percent return on your investment on ideas.

The Creativity Dice

When approaching the idea generation process it can be tempting for a team to go with what’s worked in the past and get locked into what appears to be working. Games that challenge the status quo and challenge teams in ways they don’t expect can be especially effective when generating ideas. 

The Creative Dice is a brainstorming technique that encourages participants to work in three minute bursts and work on either specification, investigation, ideation, incubation, Iteration or integration based on a roll of the dice. By moving between different modes, this method prevents premature closure of one line of ideation and keeps the session energized and engaging. What’s more, the non-linear thinking can help with idea development too!

The Creativity Dice   #creativity   #problem solving   #thiagi   #issue analysis   Too much linear thinking is hazardous to creative problem solving. To be creative, you should approach the problem (or the opportunity) from different points of view. You should leave a thought hanging in mid-air and move to another. This skipping around prevents premature closure and lets your brain incubate one line of thought while you consciously pursue another.

Affinity Map

Using brainstorming techniques to get a large number of ideas together quickly and efficiently is a great first step to developing new solutions or solving problems. But what to do once you’ve generated lots of ideas and want to work on idea development? Affinity Map is a great method for organizing your group’s brainstormed ideas and for both seeing and challenging existing patterns.

Starting with a simple brainstorm, Affinity Map asks that the group collectively organizes the ideas into columns or groups based on relationships. By doing this idea clustering as a group, your team can take ownership of the idea generation process and discover patterns of thinking together! It’s a great way of identifying and improving a group’s natural inclinations while also creating meaningful ideas.

Affinity Map   #idea generation   #gamestorming   Most of us are familiar with brainstorming—a method by which a group generates as many ideas around a topic as possible in a limited amount of time. Brainstorming works to get a high quantity of information on the table. But it begs the follow-up question of how to gather meaning from all the data. Using a simple Affinity Diagram technique can help us discover embedded patterns (and sometimes break old patterns) of thinking by sorting and clustering language-based information into relationships. It can also give us a sense of where most people’s thinking is focused

Fun brainstorming games

Brainstorming is often a fast-paced and engaging process that results in a group having fun. Creative brainstorming games that help participants have fun while generating ideas are also effective ways of loosening folks up and getting into new ways of thinking. If you’re finding your group stuck

In this section, we’ll look at brainstorming games that intentionally take a fun angle as a means to create better ideas.

Energy, fun, and creativity go hand-in-hand, and brainstorming techniques that encourage these items and generate ideas quickly and effectively – especially with large groups!

In MindSpin, teams of 3-5 participants are challenged to write as many ideas as they can in two five-minute rounds. Whenever a person writes an idea, they slam it down on the table. If they cannot think of one, they can take an idea from the person on their left and hopefully be inspired to write an additional card they also slam on the table. Remember that this brainstorming game is designed to be fast and loud while getting creative juices flowing. Encourage all participants to really slam their ideas down and keep things moving!

MindSpin   #teampedia   #idea generation   #problem solving   #action   A fast and loud method to enhance brainstorming within a team. Since this activity has more than round ideas that are repetitive can be ruled out leaving more creative and innovative answers to the challenge.

Figure storming

One potential difficulty when generating ideas is that not everyone feels comfortable sharing or holds back their ideas for fear of judgment. Creative brainstorming is all about removing restrictions or hesitation, and enabling your group to ideate freely – figure storming is a great method for achieving this!

Start by asking the group how a famous person, fictional character or well-known creative would approach the problem or topic at hand. You might ask how Albert Einstein, Elizabeth Bennett or Barack Obama might think about the ideas or concepts at hand. By inhabiting a different person’s perspective, not only can teams and individuals access new ideas, but they can also do so free of judgment. Also, it can be great fun to invite Atilla the Hun or Cleopatra to your creative exercises!

Flip and Rip

Images can often unlock our creativity but when using them to generate ideas, it can sometimes be difficult to know where to start. This visual brainstorming technique places some rules on how a group will both source and use images, and can encourage some really creative ideas!

Start by giving participants two magazines or image sets each. Then, invite each person to tear our whatever images stand out to them or grab their attention. Next, give them a problem statement and encourage them to find the connections between the problem and their chosen images. These unexpected connections between visuals can then be used as the foundation for further ideas! Plus, who doesn’t love the sound and feeling of tearing paper!

Flip and Rip   #idea generation   #problem solving   #creativity   #online   Creativity through pictures and images

Forced Connections

An important part of the ideation process is giving your team permission to be silly and bring ideas to the table without fear of judgement or inhibitions. Empowering your team to be creative without limiting themselves can massively affect the effectiveness of your brainstorming and so it’s worth spending time to unblock your participants early.

Assemble a collection of random objects or images and invite participants to choose two or more items and brainstorm how they might be used together or connected in some way. Encourage out-of-the-box thinking and unusual ideas by bringing a collection of odd items to the table and you’ll be surprised at all the ideas your team can generate!

The Thing from the Future

Science fiction and speculative thinking about the future has long been a great source of ideas. In this brainstorming exercise, invite groups to co-design their ideal future by creating tangible objects with their imaginations.

Begin by sourcing a heap of prototyping materials and craft supplies. Invite participants to imagine an ideal future state and create an object that has time travelled back to the present. After spending some time creating strange and wonderful objects, participants then present them to the group and tell stories about the objects to inform future strategies and ideas.

If you’re looking for a fun, practical exercise to bring to your brainstorming session and encourage creative thinking, this activity is a great choice!

The Thing from the Future   #imagination   #storymaking   #idea generation   #issue analysis   Help a group to time-travel and tap their imagination by fictional objects. With tangible objects and the stories your participants make up w/ them you’ll get so much richer inputs and context to inform joint visioning / strategizing: The future doesn’t look that far away when you can pick it off the shelf.

Making Lemonade

Sometimes, the best ideas come from moments of adversity or struggle. This brainstorming exercises leverages the power of positive thinking and attempting to make the best of a bad situation to generate creative ideas.

Start by sharing a couple of negative scenarios that might radically change something in the world. For example, everyone in the world has a cold, it rains constantly or we lost all our customers over night.

Next, ask your group to take something that appears negative and aim to reframe it to be as positive as possible while coming up with ideas of how that change might actually have a positive impact. Rain every day? That might be the end of drought and a boom to umbrella sales!

Making Lemonade   #creative thinking   #creativity   #design   #idea generation   Try on a relentlessly positive, can-do attitude before tackling the big stuff. The proverb goes “If life gives you lemons, make lemonade.” Practice the art of positive thinking to unlock creative ideas. Use this as a warm-up before brainstorming or to energize your team meetings.

Visual brainstorming techniques

Visual brainstorming is a great way of helping your teams out of creative roadblocks and encouraging fresh ideas. When words fail, images can enliven, invigorate and inspire your process. In this section, we’ll look at some great brainstorming techniques that focus on drawing or creating visual responses when ideating. Let’s take a look! 

Brainstorming shouldn’t stop when you have your first good idea. Fast iteration and refinement can help your group discover better ideas and develop ideas in record time. With 6-8-5, you can encourage your team to move from brainstorming to idea development in just a few minutes, and by asking participants to draw, you can keep things flowing easily!

Start by inviting your group to sketch 6-8 ideas in response to a central question or topic in five minutes. Encourage your team to be rough and not to worry about finesse at this stage – remember that the first stage of brainstorming works best when it’s free and unrestricted!

After a quick round of presenting ideas, go through the sketching process again and encourage idea development or ideas that have come out of group sharing. The result will be a heap of ideas and sketches you can move forward with too! 

6-8-5   #gamestorming   #idea generation   Part of the reason we end up with under-developed ideas is that we stick with the first good idea we have — rather than taking the time to explore complementary approaches. 6-8-5 is designed to combat this pattern by forcing us to generate lots of ideas in a short period of time. The activity can then be repeated to hone & flesh out a few of the best ideas.

Four Step Sketch

Visual brainstorming techniques can be great right at the start of the process but they can be equally effective later on when it comes to idea development. In this exercise pulled from the design sprint playbook, take your group through a structured ideation process that encourages reflection, quick sketching and a completed idea too.

Begin by reviewing any existing materials or outputs from earlier exercises before then having your group do a round of Crazy 8’s, where they create eight sketched variations on their idea. At this stage, you then invite participants to finesse their idea and create a final polished sketch to share with the team. By mixing reflection, ideation and development, this brainstorming technique offers a structured path towards better ideas!

Four-Step Sketch   #design sprint   #innovation   #idea generation   #remote-friendly   The four-step sketch is an exercise that helps people to create well-formed concepts through a structured process that includes: Review key information Start design work on paper,  Consider multiple variations , Create a detailed solution . This exercise is preceded by a set of other activities allowing the group to clarify the challenge they want to solve. See how the Four Step Sketch exercise fits into a Design Sprint

Drawing Together

Visual thinking is a powerful tool for any creative process or brainstorming workshop. This brainstorming game asks teams to tell stories about a personal or group challenge personal by expressing themselves with just five simple symbols drawn on paper. After a first draft, participants are then invited to refine their stories with colour, size and placement before the group interprets them.

By expressing themselves in a novel way, participants can exercise their creative muscles and consider new ways to express ideas nonverbally.

Drawing Together   #skills   #liberating structures   #visual methods   You can help people access hidden knowledge such as feelings, attitudes, and patterns that are difficult to express with words. When people are tired, their brains are full, and they have reached the limits of logical thinking, you can help them evoke ideas that lie outside logical, step-by-step understanding of what is possible. Stories about individual or group transformations can be told with five easy-to-draw symbols that have universal meanings. The playful spirit of drawing together signals that more is possible and many new answers are expected. Drawing Together cuts through the culture of overreliance on what people say and write that constrains the emergence of novelty. It also provides a new avenue of expression for some people whose ideas would otherwise not surface.

Brainstorming games for warming up

Generating ideas and thinking creatively isn’t always easy. Finding ways to energize and prepare your group for brainstorming with simple and fun warm-ups can help ensure the success of the workshop or session and introduce key idea generation concepts too.

While these creative games don’t necessarily create ideas in themselves, they are a great way to get teams ready for the next step. Let’s take a look!

When it comes to warming people up for an idea generation workshop, simple is best. With Draw Toast, you can introduce your group to visual thinking, working memory and systems thinking, all through the prism of a simple warm-up that works well online or offline!

For this creative exercise, invite your participants to illustrate how to make toast with a sketch or diagram without using any text. Afterward, share observations and insights as a group and outline the fact that there is no right or wrong diagram, and that differing and unique approaches to a problem or concept are all valid and useful. Not only is this a fun, fast brainstorming technique, but it prepares a group for the next stages of idea generation too!

Draw toast   #problem solving   #opening   #design   #gamestorming   #idea generation   You can use the Draw Toast exercise to introduce people to the concepts of visual thinking, working memory, mental models and/or systems thinking. This also works as a nice warm-up exercise to get people engaged with each other and thinking visually. Plus, it’s fun!

The Paper Clip Method

Even established and successful groups have preconceptions about brainstorming or how to generate ideas. Warming up a team by introducing the power of brainstorming practically can prevent stumbling blocks later on and ensure everyone is on the same page when approaching the brainstorming process.

The Paper Clip Method is a creative exercise that challenges your group to brainstorm all possible uses for a paper clip. After some silent brainstorming, debrief by sharing what the group has come up with and highlight how the group has cumulatively come up with a greater quality and quantity of ideas than any single person could.

The paper clip method   #sharing   #creativity   #warm up   #idea generation   #brainstorming   The power of brainstorming. A training for project leaders, creativity training, and to catalyse getting new solutions.

Apple-Drawing Ideation

Learning how to approach creative brainstorming as a group is a worthwhile way to kick-off your idea generation workshop. In this simple creative game, split your participants into groups and invite them to take it in turns to fill a grid with drawings of apples, with the challenge that no apple can be the same as another. 

When the time is done or the grid is full, debrief and highlight some of key concepts for effective brainstorming including the fact quantity is a precursor to quality and that we should try to build on the ideas of others. This brainstorming game is flexible by design  and you can use it with something other than apples for a more practical application: e.g., draw 30 logos, write 30 taglines, draw 30 new cars.

Apple-Drawing Ideation   #hyperisland   #innovation   #idea generation   The purpose of this simple exercise is to demonstrate three key principles useful for creativity and idea generation: quantity is a condition for quality; building on the ideas of others; the ideas we come up with are usually all the same. The format is simple, with small groups standing and drawing apples. At the end of the exercise, the whole group reflects and draws out learnings and reflections.

Activities to support a brainstorming workshop

We’ve covered everything from fast and fun creative games to in-depth brainstorming techniques, but what about if you’re looking for ways to improve or kick start your brainstorming process? Preparing for a brainstorming workshop with research on inspiration collection can really help your team make the most of the upcoming session.

In this section, we’ll look at some techniques and methods you can use to inform, enable and improve your group brainstorm and bring better quality ideas to the table! 

Rapid Research

Like all effective workshops, idea generation sessions generally want to limit the number of participants in the room to ensure productivity. That said, ideas and insight can come from anywhere in an organisation and this brainstorming technique can tap into the expertise of people outside of the room.

Start by asking participants to think of a colleague or peer outside the room and call them to get a perspective on the topic or question at hand. For remote teams, you can arrange quick calls or use Slack or other tools to get quick and useful feedback. After collecting input, bring the group back together to share and document insights. You’ll be surprised at how quickly and effectively you can get a wide variety of useful information!

Rapid Research   #hyperisland   #innovation   #idea generation   A simple exercise that complements exploratory, discursive, and creative workshops with insights and opinions from outside. Use this exercise when brainstorming ideas, developing a new product or service or creating a strategy or plan that will include others. Participants phone a co-worker and ask them questions relevant to the task. This quickly generates meaningful input from a range of “outside” perspectives. Often, participants will be surprised at how simple it was to solicit this input and how valuable it is to the process.

Lightning Demos

No idea is wholly original. Brainstorming is all about taking existing ideas and learning from others while also bringing our own insights to the table.

With this activity, you and your group will look for inspiration from how other organizations, products or thought leaders have solved or approached the problem or topic at hand. Invite participants to spend a little time gathering 2-3 examples before then presenting ideas back to the group. By the end of the exercise, you’ll have a set of ideas you can use as the basis for further discussion or idea development.

If you’re working online, collect demos in an online whiteboard and even explore the possibility of gathering inspiration before the workshop if your team has a busy schedule.

Lightning Demos   #design sprint   #innovation   #idea generation   This is an exercise to inspire your team with products or services that they think they can use as inspiration for their concepts in the next phases of their design sprints.

The Medici Effect

Great ideas can often come from sources outside of your own industry or field. The Medici Effect is inspired by Frans Johansson’s book The Medici Effect, which explores how game-changing ideas and breakthroughs can occur when concepts and ideas from one area are applied to another and used as the basis for innovation.

In this exercise, invite participants to find examples of how individuals from other fields have achieved their goals and come up with great ideas. Think of successful scientists, creatives, business owners, musicians, entrepreneurs, educators etc. have found success and what inspiration you and your team might take from them. If working online, encourage participants to include images, links and quotes so you can create an inspiration wall for further exploration of the problem you are trying to solve!

Brainstorming session templates

Are you looking to run a workshop or meeting where you need a complete group process to come up with ideas and turn those into actions? 

Check out the brainstorming session templates below to see how to build upon an initial brainstorming session with appropriate idea selection and prioritization tools to arrive at sound group decisions.!

Ideation Workshop

In this one-day workshop template, follow an entire ideation process from start to finish, going from brainstorming and idea generation through to idea development and pitching. Tap into concepts such as disruptive cases, future tech & trends and opposite thinking to create innovative ideas and empower your team!

One Hour Brain Sprint

Have imited time for group brainstorming? Try the One Hour Brain Sprint to generate ideas quickly and effectively, all while avoiding unproductive discussions and the pitfalls of some brainstorming approaches.

One-hour Brain Sprint

Remote Problem Solving Workshop

Want to solve problems with your remote team? This virtual workshop template includes several stages of ideation and development and provides a great example of how you might utilize lightning demos and research as a basis for experimental ideas and solutions.

Remote problem solving workshop

What are the benefits of group brainstorming ?

While the primary measure of success for a brainstorming workshop will be the quantity and quality of ideas generated, the benefits of this kind of session can go much further.

Establishing an ideation mindset and encouraging creative thinking will benefit your organization in the long term, and finding new ways to push your team in the direction of generating effective ideas has positive effects for your whole organization. Let’s see some of the benefits that can come from bringing team members together for a brainstorming session.

Encourage creativity 

Creative games and exercises can yield instant results when it comes to creatively engaging a team and generating ideas but beyond that, regular brainstorming can help participants be more creative in their regular work and find methods of finding new ideas and solutions that work for them. Being creative is a wonderful way to engage a group and getting out of a regular workflow can be the key that unlocks innovation.

Inclusive, easy-to-understand activities

Brainstorming is a simple group activity that is easy to understand and contribute to. Whatever skill level or competency a person has, the first stage of the idea generation process is something that can be involved in with little overhead or difficulty. This can have massive value in helping a team come together toward a shared goal in an inclusive and simple way!

Diverse ideas

Relying on certain teams or individuals to generate ideas alone can lead to stagnation. By pooling together a diverse group of people to contribute to generating and developing ideas, brainstorming can be a great way to find innovative approaches and diverse ways of thinking. Every point of view you bring to the table is another way of approaching the issue and the results generated by diverse groups are often more robust and multi-facted than those made in a silo.

Quantity of ideas

When it comes to brainstorming techniques, quantity often comes before quality. In order to find great ideas, a group first needs to flush out as many ideas as possible and share before moving onto idea development. The best brainstorming exercises encourage the creation of large amounts of ideas in a short period of time, providing a great foundation for the next steps! 

Get past creative blocks

Problem solving or idea generation can go around in circles if a team isn’t given the freedom to think creatively and approach things from a new angle. Brainstorming methods like those featured here are great ways to unblock a team’s creative and find new ways to approach stalled conversations.

Improve team morale

Brainstorm sessions are often fun and energetic by their nature, and games and exercises that focus on idea generation allow for everyone to contribute and feel heard as part of their team. These kinds of idea generation activities can really help bring a team together and improve team morale too – everyone wants to take part in developing new ideas and being creative!

Get project buy-in

Involving participants across departments and specialties early in the process by inviting them to contribute to generating and developing ideas can not only lead to great ideas but also ensure that a project is followed through on. Get buy-in early by involving stakeholders in early brainstorming sessions and help that creative energy continue throughout your project!

Kickstart projects with energy

The opening stages of a project can determine the tone for the rest of it, and by kicking off your projects with a fun, energetic brainstorming workshop, you can ensure everyone is energised for the work ahead. Try creative games to help your team approach the project with a sense of creativity and experimentation and use brainstorming techniques that see proven results to help move a team forward effectively.

Brainstorming sessions made simple

An effective brainstorming session means creating a balanced agenda of activities and group discussions while keeping everyone engaged.

With SessionLab, you  drag, drop and reorder blocks  to build your step-by-step agenda.

Your session  timing   adjusts automatically  as you make changes and when you’re done, you can  share a beautiful printout  with your colleagues and participants.

Explore  how facilitators use SessionLab  to design effective workshops that create results or  watch this five minute video  to see the planner in action!

Over to you

Brainstorming can be simple or complex, visual or virtual, but whatever method you choose to use, the results should be the same – great ideas. Finding new ways to facilitate innovation is something we’re passionate about here at SessionLab, and we hope you find the above brainstorming techniques useful! 

Did we miss anything? Are there any great brainstorming or idea generation methods you’d like to add? We’d love to hear from you in the comments.

James Smart is Head of Content at SessionLab. He’s also a creative facilitator who has run workshops and designed courses for establishments like the National Centre for Writing, UK. He especially enjoys working with young people and empowering others in their creative practice.

Leave a Comment Cancel reply

Your email address will not be published. Required fields are marked *

Going from a mere idea to a workshop that delivers results for your clients can feel like a daunting task. In this piece, we will shine a light on all the work behind the scenes and help you learn how to plan a workshop from start to finish. On a good day, facilitation can feel like effortless magic, but that is mostly the result of backstage work, foresight, and a lot of careful planning. Read on to learn a step-by-step approach to breaking the process of planning a workshop into small, manageable chunks.  The flow starts with the first meeting with a client to define the purposes of a workshop.…

Effective online tools are a necessity for smooth and engaging virtual workshops and meetings. But how do you choose the right ones? Do you sometimes feel that the good old pen and paper or MS Office toolkit and email leaves you struggling to stay on top of managing and delivering your workshop? Fortunately, there are plenty of great workshop tools to make your life easier when you need to facilitate a meeting and lead workshops. In this post, we’ll share our favorite online tools you can use to make your life easier and run better workshops and meetings. In fact, there are plenty of free online workshop tools and meeting…

How does learning work? A clever 9-year-old once told me: “I know I am learning something new when I am surprised.” The science of adult learning tells us that, in order to learn new skills (which, unsurprisingly, is harder for adults to do than kids) grown-ups need to first get into a specific headspace.  In a business, this approach is often employed in a training session where employees learn new skills or work on professional development. But how do you ensure your training is effective? In this guide, we'll explore how to create an effective training session plan and run engaging training sessions. As team leader, project manager, or consultant,…

Design your next workshop with SessionLab

Join the 150,000 facilitators using SessionLab

Sign up for free

• Skip to main content
• Skip to primary sidebar
• Skip to footer

Additional menu

Mastering the ideation process: a comprehensive guide to generating creative ideas

July 20, 2023 by MindManager Blog

In this article, we dive into the world of ideation to provide a concise and informative overview of the process. We define ideation, offer valuable tips, and explore potential outcomes. Whether you’re a professional or a beginner, this blog is a helpful resource to enhance your ideation skills. 

What is ideation?  

Ideation is the process of developing ideas or concepts. It involves brainstorming , creative thinking, and problem-solving to explore possibilities and uncover new solutions.   

Ideation is a vital step in the creative process. It is often used in design, marketing, and product development to boost innovation and generate breakthrough ideas. 

Successful ideation involves encouraging diverse perspectives and using alternative techniques to promote creativity and introduce unique and valuable ideas. 

What is an ideation session?  

An ideation session is a collaborative gathering where participants generate and explore creative ideas. It typically involves a facilitated discussion or workshop focused on a specific topic or challenge.   

Participants should include individuals from different backgrounds, such as team members, stakeholders, subject matter experts, or even end-users. 

During an ideation session, participants use various techniques to stimulate idea generation, such as brainstorming, mind mapping , role-playing, or visualizing concepts.  

The session encourages open and free-flowing discussion, allowing participants to share their perspectives, insights, and creative thoughts. It is common for ideas to build upon each other, leading to a dynamic and immersive process. 

An ideation session’s desired outcomes are to generate diverse ideas and uncover unique insights. The aim is to explore possibilities, challenge assumptions, and think outside the box.  

An ideation session encourages creativity, collaboration, and the exploration of uncommon solutions. The ultimate goal is to inspire innovative ideas that can be developed or refined to address a specific problem, improve existing processes, or create new possibilities. 

Three types of ideation processes   

When it comes to ideation, there is no one-size-fits-all approach. The beauty of the ideation process lies in its versatility, allowing individuals from various roles and industries to harness their creativity and generate practical ideas. 

In this section, we’ll explore some of the popular techniques and processes for ideation, equipping you with the tools to navigate and adapt your approach based on your unique goals. 

Three popular ideation processes include: 

1. Product ideation process  

Product ideation involves generating, developing, and refining ideas for new products or improvements.  

It starts with understanding the problem through market research and customer analysis. Then, the ideation phase generates diverse ideas from brainstorming, feedback, and trend analysis.  

Then, the best ideas are selected and transformed into concrete concepts with defined features, target audiences, and business models.  

Finally, concepts undergo evaluation, refinement, and iterations. The goal is to create a product that solves the problem, aligns with business goals, and meets customer needs. 

2. Content ideation process  

Content ideation is a creative and strategic way to generate, develop, and refine ideas for various forms of content, such as:  

• Articles 
• Blog posts 
• Videos 
• Podcasts 
• Social media posts  

The content ideation process involves understanding your target audience, generating a broad range of ideas through research and analysis, and screening and prioritizing them based on relevance and value.  

Selected ideas are developed into detailed outlines, reviewed, and refined before content creation. Factors like SEO, audience engagement, and media channels are considered.  

In addition, regular evaluation and feedback guide future improvements. 

3. Design ideation process  

Design ideation is an inventive and iterative approach to developing ideas for new products, interfaces, or visual elements.  

It starts with understanding design pain points through research and analysis. Ideation involves generating various design ideas through sketching, brainstorming, mood boards, and exploration.  

The best ideas are then developed into detailed concepts like prototypes, undergoing evaluation, testing, and improvements.  

Collaboration and feedback from stakeholders, designers, and users are vital throughout the process to ensure the final design meets the objective. 

How to run an ideation session in 6 steps 

One of the main goals of a session is to move from ideation to action. To do so, you should: 

• Establish regular updates and review sessions to track the progress of the chosen ideas. 
• Provide resources and support to overcome obstacles during the session. 
• Celebrate successes along the way to maintain momentum and motivation. 
• Continuously assess and refine the action plans based on feedback and insights. 

If you’re looking to run your ideation session, you can follow these simple six steps: 

Step 1: Define the objective and scope: 

To start, clearly define the objective of the ideation session and its scope. Identify the problem or challenge you want to address and communicate it to all participants. Set specific goals and outcomes for the session. 

Step 2: Gather a diverse group of participants 

Invite a diverse group to participate who can provide different perspectives and expertise related to the objective. This can include team members, stakeholders, subject matter experts, and end-users.  

Ensure that everyone understands the purpose and expectations of the ideation session. 

Step 3: Cultivate a comfortable environment 

Set up an environment that enables creativity and collaboration. Arrange comfortable seating and provide materials like sticky notes, markers, and whiteboards. 

Consider using visual aids or inspiring images to stimulate creative thinking. 

Step 4: Facilitate idea generation 

Introduce brainstorming methods or other ideation techniques to spark idea generation.  

Encourage participants to share their thoughts openly, without judgment or criticism, and allow participants to build upon each other’s ideas and explore different perspectives.  

Capture all ideas using visual tools or digital platforms. 

Step 5: Evaluate and prioritize ideas 

Once a good number of ideas have been generated, facilitate a discussion to evaluate and prioritize them.  

Ask participants to consider factors such as feasibility, impact, goal alignment, and risks. Use voting or ranking methods to identify the most promising ideas. 

Step 6: Develop actionable plans 

After selecting the best ideas, develop actionable plans and transition from ideation to action.  

Assign responsibilities, set deadlines, and define specific steps for each idea. Create a roadmap that outlines key milestones, resources, and potential challenges. 

Ideation process example  

The ideation process looks different depending on what you’re brainstorming. For example, let’s pretend we’re looking to conduct an ideation session for a new mobile app.  

Below is how ideation can happen before, during, and after your session: 

• Before ideation. Conduct market research to understand user needs, competitors, and trends. Identify target demographics, pain points, and app objectives.   
• During ideation. Assemble a cross-functional team. Define problem statement and outcomes. Brainstorm ideas using techniques like storyboarding and sketching. Foster collaboration, challenge assumptions, and evaluate ideas based on feedback and feasibility.   
• After ideation. Evaluate and select promising concepts and create prototypes. Conduct usability testing and refine the design based on feedback. Once completed, move to app development. Continuously monitor performance and implement updates based on user needs and trends. 

Ready to boost your ideation sessions? MindManager is a mind mapping tool that can help you and your team brainstorm.  

Get started now and explore the power of mind mapping for generating innovative ideas and enhancing collaboration. Elevate your brainstorming sessions and unlock the full potential of your team.  

Sign up for a free trial and supercharge your ideation process with MindManager! 

Ideation process frequently asked questions (FAQs)  

Below are some of the most commonly asked questions about the ideation process: 

Ideation is the process of generating creative ideas and concepts to solve problems or explore new opportunities. It involves brainstorming, exploring different perspectives, and thinking outside the box to develop creative solutions.  

Ideation is a critical phase in the design and innovation process, serving as a foundation for further development and refinement of ideas. 

What are some different types of ideation?  

There are several different types of ideation techniques depending on the context and desired outcomes. For example: 

• Brainstorming. A group activity where participants freely generate ideas and suggestions without judgment or evaluation.   
• Mind mapping. A visual technique that involves creating a diagram or map to organize and explore connections between ideas visually. 
• SCAMPER. An acronym-based technique that prompts individuals to consider different approaches by substituting, combining, adapting, modifying, putting to another use, eliminating, or reversing elements of a concept. 
• Reverse ideation. Focusing on identifying problems or obstacles instead of solutions to stimulate creative thinking. 

What is an ideation session?  

An ideation session is a structured and collaborative gathering where individuals or teams gather to generate creative ideas and solutions for a specific problem or challenge. It involves brainstorming, open discussion, and exploring various perspectives to foster innovation and generate various potential ideas. 

Ready to take the next step?

MindManager helps boost collaboration and productivity among remote and hybrid teams to achieve better results, faster.

Why choose MindManager?

MindManager® helps individuals, teams, and enterprises bring greater clarity and structure to plans, projects, and processes. It provides visual productivity tools and mind mapping software to help take you and your organization to where you want to be.

Explore MindManager

7 idea generation methods for out-of-the-box thinking

By Anabelle Zaluski

Successful organizations run on great ideas. 

Whether you want to improve standard protocols or pitch original initiatives, fresh ideas keep your team and projects from growing stagnant. But idea generation sessions can get messy, with disparate thoughts or unfocused tangents stealing focus from more pressing issues. 

Luckily, various techniques exist for encouraging focus and more innovative solutions. Find the right method for your team and enjoy a tried-and-true way of turning out-there concepts into actionable innovation. 

Idea generation definition: What is it, and why is it important?

Idea generation — or ideation — is the collaborative creation and development of ideas. It kicks off the idea management process, through which you can build out abstract, concrete, or visual concepts to determine solutions or explore new opportunities. 

Fostering a culture that welcomes fresh ideas allows teams and organizations to grow with market shifts and fluctuating consumer demands. When a marketing team brainstorms a new campaign or product managers scout expansion opportunities, diverse perspectives in the ideation stage empower innovation. Varied viewpoints enrich the proposal pool and encourage innovative insights, paving the way for more strategic and successful strategies and initiatives. 

How a proactive idea generation process benefits your team

Creative thinking turns vague ideas into groundbreaking changes, turning your team into industry leaders. Here are a few more pointed ways effective idea generation benefits your team: 

Offers a competitive advantage — The right ideation method for your team brings out the best in them, resulting in initiatives that help you stand out in the market. Getting creative with brainstorming techniques doesn’t mean you have to act on every wild idea. It’s about setting the stage for forward-thinking concepts. And when an innovative, worthwhile idea pops up, you’re ready to refine it into a product or service that resonates with consumer needs and market trends. 

Enhances team engagement and morale — effective idea-generating techniques bring all stakeholders on board, making team members feel vital to the company’s mission. And when insights transform into actionable strategies, it fosters collaboration, productivity, and more open communication.

Enables anticipatory problem-solving — being proactive with regular solution generation lets you anticipate challenges rather than scramble for solutions. Consistent brainstorming focused on improvement pinpoints potential hurdles early, leaving room for preemptive problem-solving. This forward-thinking approach facilitates strategy adjustments and lets you capitalize on emerging opportunities. 

7 idea generating techniques to try with your team

Depending on your objectives, here are seven idea generation methods to tease innovative concepts out of your team.

1. Mind mapping

Mind mapping is more than just a visual outlining method. It’s a structured approach to organizing complex information and concepts. 

Imagine you’re launching a new product. To create a mind map, you’d write the product's name in the center of your outline. From there, you’d draw branches representing major categories, such as marketing, product design, and distribution. Next, you’d add sub-branches that detail specific actions or elements. 

This visual representation focuses on organizing every project detail, highlighting interdependencies, and ensuring teams see the big picture so they can collaborate effectively. It’s a useful way to plan complex projects while keeping sight of overarching objectives. 

2. Blue sky thinking/brainstorming 

Blue sky thinking, or brainstorming, encourages unrestrained creativity. When developing a new project, team members throw out specific details or constraints and voice any and every idea, no matter how unconventional. 

While the approach might sound chaotic, focusing on quantity rather than the quality lets groundbreaking concepts emerge. Once brainstorming concludes, you scrap unrealistic suggestions and turn viable ones into actionable strategies aligned with your product vision board .

3. Brainwriting

Like brainstorming, brainwriting focuses on amassing different perspectives. But rather than shouting out ideas, team members jot everything down on paper. They then pass their papers around in a circle, and receiving teammates build on each other’s ideas by adding new perspectives or pointing out pain points. The cycle continues until everyone has their original paper back — now fleshed out with team reactions and elaborations.

This technique diversifies ideas and stimulates inclusivity, which is useful for large groups where some team members might be more comfortable verbally sharing ideas than others. 

SCAMPER stands for substitute, combine, adapt, modify, put to another use, eliminate, and reverse. It’s a technique to reimagine existing ideas, services, and products. For teams aiming to refresh their offerings, SCAMPER is a structured way to dream up enhancements and new directions. 

Imagine a product development team brainstorming ways to revitalize an old product to meet current consumer trends. Using the SCAMPER method, they explore multiple options for reinvention. The “substitute” prompt might lead to swapping a material for another with a lower cost, while “adapt” might stimulate ideas for tweaking the product to cater to a new audience. 

5. Role-playing

Sometimes, it’s helpful to step out of your role as a designer, developer, or project manager and think from your consumer's perspective. Role-playing requires you to simulate the users you aim to serve. 

Think about building a mobile app. Instead of relying on market data alone, role-playing would have you consider the app from the perspective of your target audience. You might think of how busy parents and teenagers uniquely navigate the app, providing real-time feedback to generate potential refinements and spotlight user concerns. 

If your team struggles to reverse roles, build AI-generated scripts based on customer metrics to create an accurate back-and-forth.

6. Storyboarding

Like frames in a comic strip, storyboarding sequences your idea's journey from concept to final product. Using images, quotes, and other graphics brings procedural or product management to life. Storyboarding transforms abstract ideas into clear visual narratives, pointing out potential areas for enhancement and friction points in a way that suits visual learners. 

7. Synectics

Teams sometimes get stuck on existing thought patterns, struggling to look beyond solutions they’ve already discussed or executed. Synectics utilizes unrelated problem analogies to stimulate creativity and generate new ideas.

Consider a development team that misunderstands customer needs and wastes time and resources designing unnecessary functions. The team could liken their situation to buying expensive toys for a dog who doesn’t like to play with them. The team would then brainstorm ideas to fix the issue in the analogy — like letting the dog choose a toy at the pet store. This idea can inspire the team to approach their problem in a similar way — by letting the consumer dictate their own preferences rather than prescribing them. 

To put this idea into action, the team could generate customer surveys or lead focus groups to gather user insights and then refine their development approach accordingly.

Tools for managing ideas

You experimented with several idea generation techniques and now have many concepts to iron out. Without proper management, potential catalysts for innovation might get lost. Arming yourself with effective tools and alignment strategies ensures that change is innovative and purposeful. 

Templates and management tools let you store all your ideas in one place. A shareable template ensures all stakeholders can revisit and organize thoughts in a centralized space. And materials like vision-to-values boards let teams measure the value of an idea against the organization’s overarching mission and goals. 

3 tips for successful idea generation sessions

Solution generation requires more than just choosing a method. It’s up to managers to set the scene and create guidelines to push brainstorming in the right direction. Here are three tips to consider.

1. Clearly define the problem or opportunity

Ideas that successfully tackle a problem or seize an opportunity need clear frameworks. Start by giving a detailed definition of what ideation seeks to accomplish. When team members understand the core issue, it directs brainstorming efforts and ensures targeted and relevant suggestions.

2. Involve the right people

Idea generation benefits from diversity of thought, experience, and expertise. To enrich your solutions pool, welcome insights from team members across various departments and roles. Incorporating varied viewpoints helps you anticipate cross-departmental challenges, identify overlooked opportunities, and craft well-rounded solutions. 

3. Consider execution 

Throwing out ideas is just the beginning of successful ideation. It’s just as essential to develop a clear plan for evaluating, refining, and implementing new initiatives. Tools like Notion’s idea-to-execution template help nurture suggestions from start to finish. 

From brainstorm to breakthrough with Notion

Idea generation is a fun — and oft-chaotic — team exercise. But with structured tools, you can turn out-of-the-box thinking into actionable strategies. 

Notion’s idea generator and product and content idea generator templates enrich the process by structuring your process. Or search the template gallery to find something that perfectly suits your needs.

Keep reading

Why AI gets better when we let it explore

Cultivate collaboration with communities of practice

Turn ideas into reality with an agile engineering design process

Team synergy and its importance in project management

Problem-solving: Steps, strategies, and skills

Top 9 brainstorming techniques to drive innovative thinking

Get going on web or desktop, we also have mac & windows apps to match., we also have ios & android apps to match..

Desktop app

Creative Ideation: Techniques for Generating Innovative Ideas (the Complete Guide)

Jack O’Donoghue

Creative and corporate – Two words that don’t strike us as complimentary.

In this article, you’ll learn how to generate innovative ideas and level up your innovation game with creative ideation. The most innovative teams in the world have tried, tested, and proven the idea generation techniques discussed in this article.

You might also like…

• How to Use Design Thinking to Solve Problems (the Complete Guide)
• How to Run ‘How Might We’ Workshops: A Quick Guide
• How To Use ChatGPT For First Principles Thinking In UX

Table of Contents

Did you know, the power of creative ideation, what makes an idea innovative, the 4 types of innovation, methods of driving creative innovation, creative ideation and brainstorming techniques, how to generate innovative ideas.

• How to Plan and Facilitate an Ideation Workshop 

Idea Generation Checklist

Further reading.

To help teams implement Design Thinking , we offer bespoke innovation training workshops. Talk with us and find out how we can help transform the way you design your products and services.

Interested? Message us in the bottom right corner or learn more here.

What Can You Achieve With Creative Ideation?

A well-executed ideation session can be an extremely valuable tool for you and your team. It can promote creative thinking, bond your team, give everyone a sense of ownership, and bring new perspectives on challenges and problems. It can help you create a roadmap and backlog of creative ideas to propel you ahead of the competition. 

Facilitating the creation of innovative ideas is an excellent opportunity for you to demonstrate your skills as a creative leader.

Who Can Take Part in the Ideation Process?

Anyone can contribute to an idea or participate in idea creation; you can ideate with large or small groups, generate ideas on your own, or for an extra layer of insight, you can even ideate and brainstorm with customers or your target audience. In fact, the more diverse the group of people, the more diverse the ideas will be. 

The tools and techniques discussed in this article apply to any ideation scenario; they’re proven effective for generating ideas for products, features and services. The frameworks outlined below are extremely popular with the world’s best software development and marketing teams. 

What Is Creative Ideation?

Creative ideation is the process of generating, evaluating, and selecting new ideas. We use creative techniques to unlock new perspectives and help us achieve our goal of creatively solving a problem statement or challenge.

Creative ideation usually takes place in a group or workshop setting. The outcome we’re after is to produce a collection of new ideas, ready for us to test with our customers and develop before shipping to market.

What Does Great Creative Ideation Look Like?

There’s nothing quite like a well-run creative ideation session. The atmosphere should feel light and exciting; the room should buzz with conversation and enthusiasm about new ideas.

Great ideation will leave a team feeling inspired and ready to commit to achieving their goals. Great ideation will be a memorable moment and stand out to its participants as a project highlight. Great ideation can change the direction of a team and inspire stakeholders to give their buy-in and support.

Valuable Innovations Solve Real Problems

Innovations solve real problems for real people – the more people who experience the problem you’re aiming to solve, and the bigger the problem is for this audience, the more significant impact your innovation will have.

Inventive ideas that don’t solve problems won’t gain traction. It’s risky to attempt to solve problems you know very little about. It’s easy to fall in love with an idea before you fully understand the problem; the consequence is that you commit your time and resources to an idea that doesn’t add value to your target audience’s lives.

Creative Solutions vs Non-Creative Solutions

Not all ideas need to be left-field, ground-breaking disruptions. Often the most straightforward idea is the best. Simple ideas are easier to execute, take less time and resources, and fall within your existing capabilities.

However, creativity thrives under the tightest of constraints. Real innovation emerges when one’s back is against the wall, and the stakes are high. Creativity is necessary for competitive industries, saturated markets, when your product is easy to replicate, or when your team is short on time or resources.

“The simplest solution, is almost always the best” Occam’s Razor

When you’re setting out on the path to innovation, it’s helpful to understand the different types of innovation and how to leverage them within your strategy. Understanding the different kinds of innovation can help you choose a focus or a combination of methods that will give you the best chance of success. 

Here are 4 types of innovation for you to consider:

Basic Research

Basic Research is the process of continuously gathering information. Over time your knowledge bank grows, your insights deepen, and your learnings compound. Gaining this deep insight can help you iterate your product over time. Each iteration can add up to massive value. This approach is useful when the problem needs to be better defined, and you need to know your domain better and learn more about your customers and target audience.

Breakthrough Innovation

Breakthrough innovation occurs when you’ve clearly defined your problem but have yet to be able to solve it. When we combine new skills and expertise, we unlock new perspectives and achieve breakthrough innovation. Diversity of thought is a vital ingredient in the innovation process. Including experts on various subjects gives creativity a chance to connect dots you otherwise wouldn’t have been able to connect.

Sustaining Innovation

Sustaining innovation is what happens in most businesses. It’s the process of continuously improving existing products, within an established market, for a customer you’re already familiar with. It isn’t ground-breaking; you iterate and improve your product to sustain an innovation already in place.

Disruptive Innovation

Disruptive innovation is what most of us think about when we hear the word innovation. Disruptive innovation leads to the creation of new markets; it changes industries and moulds consumer behaviour. It’s when we solve problems in a way we never thought possible; it’s the holy grail often pioneered by the underdog with little to lose.

We can choose from many helpful innovation frameworks to help us develop fresh ideas. These frameworks are tested, proven, and established, ready for you to apply to your work.

Notice that each framework shares similar phases: learn, ideate, and test. 

These steps are critical to all design processes; each framework differs in context, values and intended audiences. Still, each can be valuable to you if your goal is to generate new inventive ideas. 

It’s best to take the time to understand them, try them out, and find what works best for you. Remember, these are just frameworks; if you have special requirements, you can modify and adapt the framework to meet your needs.

Here are 5 methods for driving creative innovation:

The Creative Ideation Mindset

The creative mindset requires trust in the process and our intuition. It’s a mindset that allows thoughts to flow freely without judgement. The creative mindset is about freedom of thought, curiosity, experimentation and imagination. Creative thinking loves analogies, metaphors, and similes.

We use our unconscious to stream our thoughts, combined with lateral thinking, to come up with unusual ideas before analyzing them. This process is how we create unexpected connections and realize fresh new ideas. 

An essential part of the creative process is to take breaks to allow your mind to process the information. After a certain amount of thinking, do something relaxing and unrelated to enable your thoughts to percolate. It’s often in these moments when those highly sought-after connections occur.

Human Centered Design

Human centered design is an evidence-based approach to problem solving. It prioritizes the needs of the people that we’re designing for. 

Human-centered design requires understanding your user’s goals, motivations and pain points and involving them throughout the process to help us make decisions.

Getting feedback throughout the process improves your chances of getting it right. So we create opportunities to validate our thinking by getting customer feedback at different stages of the process.

Getting feedback throughout the process improves our chances of success. It helps to guide our thinking and reduces the risk of failure. So it’s often cheaper to use a human-centred design approach because our work is evidence-based and validated, which gives us confidence before we commit our resources.

Jobs to Be Done

Jobs-to-be-done is a framework created by Clayton Christensen – a Harvard Business School Professor – credited with creating the theory of disruptive innovation.

The theory states that people hire products as tools to get their jobs done. The classic example is that people don’t buy a drill. They buy the hole in the wall. Although, one may even go further to say that what they purchase is a shelf hung securely.

Jobs can be emotional, functional, or social. We research to understand this need before generating ideas and validating them with customers.

Design Thinking

Design thinking is a human centered design approach to solving problems. 

The first step is to gain empathy for our customers by speaking with them and conducting user research. We then define the problem , identify pain points, and discover opportunities by analysing the information gathered in the previous phase. 

The next step is to generate, discuss and develop our ideas before finally prototyping and testing them with our customers.

Design Council’s Double Diamond

The Design Council developed the Double Diamond as an end-to-end framework for innovation.

The first step is to discover our subject and learn all we can through conducting research and gathering information. Then we define a problem and articulate a challenge. After that, we develop solutions and deliver the outcome.

The Double Diamond values people, communication, and collaboration as the key drivers of the innovation process.

Design Sprint

Google ventures developed the Design Sprint as a rapid idea development framework. It helps teams create new features for products and services during a 5-day process where we generate and validate our ideas.

Design Sprints are collaborative, fast-paced, and based on human-centred design principles. The first step is to map the customer experience and choose a Focus. We then generate, evaluate and discuss inventive ideas.

Finally, we choose the best ideas and build a realistic prototype. We then test this idea with customers to validate our thinking. It’s important to note that the prototype doesn’t need to be functional. It can be a sketch, a low-fidelity mockup, or even a written explanation. It’s your choice of which format is most suitable for you, depending on your time and the information you need to gather at this phase.

Brainstorming and ideation are sometimes used to mean the same thing. But there are subtle differences. A brainstorming session is an idea generation technique that we use to generate a wealth of ideas. 

Brainstorming can be done on the spot with little preparation. It can be an informal and spontaneous approach to problem solving.

Ideation is the process of generating, evaluating, and selecting ideas. 

Ideation takes planning, prep work, time and commitment.

The brainstorming technique is used within the ideation process. We’ll talk more about the end-to-end ideation process later. For now, let’s look at some creative or idea generation techniques.

Here Are 10 Creative Idea Generation Techniques

Warning! Bravery and trust are required to get the most from these idea generation techniques. The more experimental techniques can sometimes make people uncomfortable if they’re unfamiliar with the creative process. 

But these creativity techniques will make generating fresh ideas fun.

If you’re facilitating the session, it’s your job to ensure everybody feels comfortable and safe sharing their thoughts and ideas. Let everyone know that we aren’t judging their thoughts at this stage. We’re here to have fun and imagine new possibilities. Encourage everyone to trust the process and reassure them that this is a time to explore and play.

You should experiment with different techniques. Combine several, or modify them in a different way to get a different perspective and discover the best ideation technique for you.

Random Association

• Choose a challenge or a problem to solve.
• Assign each participant with a completely random word.
• Based on the random word, write down the first related word you think of.
• Now, based on the last word you wrote, think of another related word.
• Continue to write related words, in sequence, for a set amount of time.
• Now, ask everyone to choose a word from the list you just wrote.
• Think about how the random word relates to the challenge or problem.
• Now, sketch out some ideas inspired by the random connection.

Creative Matrix

• Draw a large grid on a wall or whiteboard.
• Across the top row, list your customer segments or personas.
• Down the left column, list the journey phases or key insights.
• Brainstorm ideas within the cells based on the intersecting points.
• Talk through the ideas and vote on your favorites.
• Take an existing idea, product or feature as a starting point.
• Put it on a wall or whiteboard and give everyone sticky notes.
• Look at the wall, ask questions, and capture the responses.
• Substitute : What can we substitute for something else?
• Combine : How can we combine two or more parts?
• Adapt : Which parts can we adapt or change?
• Modify : What can we modify, exaggerate or emphasize?
• Put to another use : How else could this be used?
• Eliminate : What can we remove, reduce or simplify?
• Rearrange : What can we rearrange, reorder or reverse?
• Take your reflections, discuss them, and use them to iterate the idea.

6,3,5 Brain-writing

• Choose a challenge or problem to solve.
• Gather 6 people around a table or on a video conference.
• Ask each person to take 5 minutes and brainstorm 3 ideas.
• Then pass their ideas to the next person to build upon.
• Continue passing the ideas around until you’ve completed the circle.
• Review and discuss the ideas together.
• Vote on the ones you want to take forward.
• For a similar exercise that can be done solo, see ‘mind mapping’.
• Gather several small groups of 3-4 participants.
• Each group should discuss the challenge and sketch ideas.
• The group then reviews, refines and selects the best ideas.
• Each group then passes their ideas to the next group.
• Continue passing ideas around, building on each other’s work.
• Once the circle is complete, organize, prioritize and select ideas.

Starbursting

• Choose a product, idea, theme, problem or challenge.
• Draw a 6-pointed star and write the chosen stimulus in the centre.
• On each point, write in sequence: Who? What? How? Where? When? Why?
• Brainstorm questions starting with the prompt on each point.
• For example: Who? – Who would get the most value from this product?
• Continue around the star, brainstorming questions until you’ve finished.
• Now, ask the group to vote on the most thought-provoking questions.
• Take some time to discuss and answer the questions. 
• This is an excellent exercise for kicking off an ideation session or workshop.
• Gather a group of people and split them into two groups.
• One group will discuss the challenge while the other group observes.
• Then switch so the other group discusses and the other group observes.
• When the time is up, get together and discuss the observations.
• Have someone write up notable ideas and thoughts on a board.

Round Robin

• Gather a group of people in a room or on a remote whiteboard.
• Each person is to spend some time generating solutions to the problem.
• Everyone then passes their idea to the person next to them.
• They then create an idea based on the idea they’ve just received.
• Continue going around the circle as many times as you like.
• Once the time is up, group the ideas into themes.
• Discuss the themes and vote on your favorite ideas.

Rapid Ideation

• Set a timer for 8 minutes.
• Ask everyone to generate 8 ideas.
• Ask everyone to share their ideas back to the room for discussion.
• Repeat the exercise multiple times, introducing different stimuli.
• For example: “Now imagine you’re Elon Musk; generate another 8 ideas.”
• “Now think about how a child would solve the problem.”
• Start generating ideas and writing them on a board.
• Draw a branch from each idea and write something inspired by it.
• Continue generating ideas until you’ve filled the page.
• You can do mind mapping solo or in a group. 

Below I’ve outlined the process for generating innovative ideas. These ideas come from years of experience as a design leader practicing the design frameworks we looked at before. Start at the beginning and work your way through. After repeating the process a few times and gaining experience, you can adapt and modify it based on the projects you’re working on.

Here’s an end-to-end idea generation process for discovering innovative ideas:

Choose a Starting Point

If you already know what problem you want to solve, you can skip this Step. If you don’t know where to start, or need a new angle, it’s common to feel frustrated or overwhelmed when choosing a starting point.

The design process is ambiguous and at times confusing. When we get started and move forward, things become clearer. After all, it’s at the beginning of a project when we know the least.

The best way to start is by speaking to your customers. Seek to understand their experiences, goals, needs, and pain points. You can also look at data and Analytics to help you identify a problem area to begin your investigation.

Or, if it’s a less formal project, you can choose an area of interest, dig deep, and find a worthy problem.

Speak to Your Customers

All effective design processes start and end with the customer. We first speak to customers to understand their experiences and goals. We talk to them along the way to get feedback and to help guide our thinking. We speak to them at the end of the project to validate our ideas and gain confidence in our direction.

This stage of the project is about building empathy and gaining insight into our target audience’s experiences, behaviors, and attitudes.

You can speak to your customers during one-on-one interviews, focus groups, and workshops, or you can send out surveys and poles together quantitative information. You can recruit customers using online recruitment tools, recruitment agencies, or your social media and owned channels.

Define Your Problem

After you’ve spoken with your customers, you’ll have a heap of data. You’ll need to organize this data and turn it into a useful format. Some artifacts that help to organize your data are personas, journey maps, and problem statements.

Having a clearly defined problem is critical to success. You’ll use it as a jumping-off point for ideation, to evaluate your ideas, and you’ll use it to create metrics to determine whether or not your innovation has been successful.

Take your time to get to know the problem. Add a layer of creativity by reframing your problem a few times. Reword it and look at it from different angles to see if you can inspire an unexpected direction.

 “If I had an hour to solve a problem, I’d spend 55 minutes thinking about the problem and five minutes thinking about solutions.” Einstein

Research the Competition

Before you generate ideas, it’s essential to know what the competition is up to. By competitors, we don’t just mean direct competitors within your industry. We mean any other product or service that aims to solve the same problem as you. For example, Netflix CEO Reed Hastings quipped that their main competitor was sleep. Spend some time reviewing the competition, compile a competitor analysis, and identify threats that they pose and opportunities they’ve left exposed.

“You get a show or a movie you’re really dying to watch, and you end up staying up late at night, so we actually compete with sleep.” Reed Hastings – Netflix CEO

Get Inspired

Now it’s time to put your creative hat on. Creativity feeds on inspiration. It connects dots and creates pathways. Use this time to have fun and find inspiration. Don’t just look at competitors. Look at un-related fields too. You can look to art, architecture, nature and music for inspiration. The best ideas can come from drawing analogies to unrelated fields.

Generate, Evaluate, and Select Ideas

At this stage, we want to use one of the many brainstorming techniques discussed earlier in this article. Typically at the start of ideation, we aim for quantity over quality. The aim is to get all the obvious ideas onto the table. We then encourage ourselves and our teams to push beyond them. 

For effective idea generation, we need to feel safe sharing our thoughts. 

Encouraging people to open up is the art of facilitation.

Once we’ve generated a range of different ideas, we discuss and critique them with our teams. We want to ensure that our creative solution is solving our problem. We can do this as a written or verbal exercise. A decision-making matrix can help us select ideas. Another helpful tool for vetting ideas is the DVF framework, where we assess ideas based on their desirability, viability, and feasibility.

Get Feedback on Your Idea

Finally, we get feedback from customers. It doesn’t mean it’s the last step in the process, as it isn’t linear. It’s cyclical. Often we cycle through the phases of the ideation process to deepen our thinking. We can cycle through the same phase multiple times, repeat the process on different aspects of the ideas, or go deep in some areas and stay lean and agile in others.

At this stage, we want to validate ideas and learn from customers. We want to see them interact with the idea, share their thoughts, and observe their reactions. We then use this information to develop the ideas further. 

At this point, you need to create something to share with customers. Now is where the prototype comes in. It can be a functional prototype, a sketch, or a description. It’s up to you to decide which format is necessary for your goals.

How to Plan and Facilitate an Ideation Workshop

How to Plan a Creative Ideation Workshop

The quality of your ideation workshop will directly reflect the quality of your plan. To help you with the planning phase, I’ve created a canvas for you to use to help prepare for the session. You can do this independently or work through the canvas with your team or stakeholders.

You’ll need to articulate clearly the goals of ideation and the problem you’ll be solving. And you need to have an agenda and a set of activities prepared. After, you need to send out invitations and book participants.

How to Facilitate a Creative Ideation Workshop

Facilitating an ideation workshop is about creating a safe space for your participants. People need to focus on the task without overthinking the instructions.

If people aren’t familiar with or haven’t taken part in an ideation session before, it’s common for them to feel hesitant to open up and share their thoughts. It’s your job to create a safe space and ensure everybody feels included and valued.

The participants must be absolutely clear on what they’ve been asked to do. They need to enjoy the process, be relaxed, and feel comfortable. It’s your job as a facilitator to make sure that these needs are met.

How to Keep Momentum and Gain Traction

After your ideation session, you need to keep momentum to gain traction and get your innovative idea off the ground. Sharing the process and outcomes with stakeholders can encourage them to buy into your ideas and inspire them with new possibilities. Product development requires the support of the whole organization, so consider how you can provide evidence and inspire them to buy into your great idea.

Since your ideas may not be fully formed, they’ll likely need further research and development sessions. Create a plan to continue developing your ideas. Speak with project managers to prioritize your ideas and ensure they’re giving the attention they deserve. Use the participants from the session to help gain traction, turn them into advocates, and get their support to take the ideas further.

Ideation Tools and Materials

If you’re running a remote workshop, you can do everything online. All you need is a whiteboard or collaboration tool like Miro or FigJam. Both of these have free plans that you can use for your workshop. All you’ll need to do is set up your activities on the collaboration board and include the URL in the invites you send out to participants.

If you’re running a workshop in person, you need more materials. You need pens and paper for everyone to use—a blank wall or whiteboard to capture notes and pin-up ideas. And, sometimes, a projector helps share insights and context. It’s also a good idea to plan drinks, snacks, and food. Keeping everyone fed and making them comfortable is essential.

Here is an idea generation checklist to remind you of the steps in the process and what you need to do along the way. 

Alternatively, bookmark this article and refer to it when ready to discover your new idea.

• Choose a starting point
• Speak to customers
• Gather some insights
• Research the competition
• Get inspired
• Generate ideas
• Evaluate ideas
• Select ideas
• Get feedback from your customers
• Jobs to Be Done (Know your customer’s jobs to be done)
• Double Diamond (Double Diamond: Framework for innovation)
• Design Thinking (Ideo: What is design thinking?)
• Design Sprint (Google Ventures: The Design Sprint)
• Gamestorming (Ideation games and fun activities)

How useful was this post?

Click on a star to rate it!

Average rating 3.7 / 5. Vote count: 3

No votes so far! Be the first to rate this post.

As you found this post useful…

Follow us on social media!

We are sorry that this post wasn't useful.

Let us improve this post!

Tell us how we can improve this post?

Related Articles

How Design Thinking is Addressing 5 Major Challenges in Healthcare

The Basics of Design Thinking for Improved Patient Care

8 Tips for Implementing Design Thinking in Your Healthcare Practice

The Science of Brainstorming: How to Effectively Generate New Ideas

Anne-Laure Le Cunff

Many people believe that creativity is a natural gift that only a select few are born with and that it cannot be taught or learnt. This could not be further from the truth.

Yes, creativity is innate in the sense that we are all born with it. But, as we grow up, most of us slowly unlearn it . The good news is that what is unlearned can be learned again. It’s just a matter of figuring out how.

There is a lot of content out there with various tips and tricks that may or may not work. What does the science say? How can we be more creative and effectively brainstorm new ideas?

Quantity versus quality

We have an implicit conception that good work takes time. This is why prolific authors are often judged as bad; and their work, inconsequential. In an amazing essay for the New York Times titled “Can a novelist be too productive?”, Stephen King — who has published more than 55 novels — argues that while quantity is never a guarantee of quality, being prolific can definitely result in quality work.

Agatha Christie wrote 91 books and gave us Hercule Poirot. Picasso painted over 20,000 artworks. James Dyson developed 5,127 prototypes when trying to design a better vacuum cleaner. Thomas Edison still holds the record for the most patents with over a thousand in his name. Were all of these groundbreaking? Probably not, but that’s exactly the point.

It may sound counterintuitive, but research suggests that quantity yields quality when it comes to creativity. In the book Art & Fear , David Bayles shares the anecdote of a ceramics teacher who conducted an experiment with his students. He divided the class into two groups.

Group A was to be graded based on the quality of the work they produced, whereas group B would be graded on quantity. To get a perfect grade, group A had to produce only one pot — the most perfect ceramic pot possible — while group B would have to create as many as possible.

The results are fascinating: when it was time for grading, the best work came out of group B, the “quantity” group. While group A was busy debating and theorizing, group B was dutifully creating pots after pots, and learning from their mistakes in the process.

Think you’re out of ideas? According to research , we tend to grossly underestimate how many ideas we can generate. Even more interesting, according to the same research, the more ideas we keep on generating, the more creative they become.

Creative routine

Of course it’s tempting to spend a lot of time reading and researching your area of interest — and such research also has its place! — but you will not improve your creative thinking without consistent output.

Creativity is like a muscle. You need to use it to stay in “creative shape”. This means — however uncomfortable that may sometimes be — forcing yourself to create on a schedule.

Whether your goal is to write a book, become a better illustrator, or build an app, don’t leave creativity to random bursts of inspiration. Block some time every day or every week to generate new ideas and new work. I personally use the PARI framework to ensure my daily creative output aligns with my long-term ambitions, but as long as you flex your creative muscle consistently, you will be on your way to do your best creative work.

As poet W.H. Auden so beautiful put it: “Decide what you want or ought to do with the day, then always do it at exactly the same moment every day, and passion will give you no trouble.”

Having a creative routine allows you to keep your cognitive bandwidth for creative thinking. According to William James, considered by many as the father of modern psychology, such routine allows us to “free our minds to advance to really interesting fields of action.” Basically, the resources you don’t waste trying to decide when or where to do creative work can be used to, you know, to actually do the work.

So, how can you go about creating on a schedule?

• Get up early or stay up late. There is no right or wrong way to go about your routine. Some creative people are early risers, others are night owls. In her diary , Anaïs Nin wrote: “I do my best work in the morning.” In contrast, Jack Kerouac said : “I had a ritual once of lighting a candle and writing by its light and blowing it out when I was done for the night.” Pick whichever works for you.
• Choose your creative space with intention. If you can, find a secluded spot that is solely dedicated to creating. Especially when working remotely, we tend to just sit wherever, for example at the kitchen table. Pick a spot and make it your creative space. Make it comfortable, and make sure to have all the creative tools you need.
• Warm up for creative work. Take some time to loosen up and get your creative juices flowing. Write a few sentences without thinking too much, play with your design software for a bit without trying to create something concrete. This process will give your mind time to shift into a more creative state.

If you want to read more about the creative routines of famous artists and inventors, I recommend reading Daily Rituals by Mason Currey — full of fascinating stories. Disclaimer: some routines include taking mind-altering substances and other more original approaches — which can have their place in the creative process, but as usual don’t just blindly apply what you read.

Pick your creative mode

According to psychology research , there are several types of creativity you can leverage to brainstorm ideas more effectively.

• Combinational creativity. We are often seeking original ideas, when in reality most creative concepts are a combination of old ideas. First, collect as many old ideas as possible. This can be done by reading science fiction or just taking notes every time you hear a commonplace idea in a conversation. Then, let these old ideas incubate for a while. Yes, there’s no second step. Let your brain do the work. “Drop the problem completely and turn to whatever stimulates your imagination and emotions. Listen to music, go to the theatre or movies, read poetry or a detective story,” recommends James Webb Young. The combined idea will most probably come to you when your brain is relaxed, such as in the shower.
• Exploratory creativity. In academia, exploratory creativity is defined as “the process of searching an area of conceptual space governed by certain rules.” This means that you try to generate new ideas within a given space, taking into account its specific rules. For example, let’s take transportation. Why is it expensive to fly? Why is it so hard to find a taxi? Exploratory creativity is all about exploring existing concepts and ideas you may already have and questioning their validity to come up with new solutions.
• Transformational creativity. This method takes things even further. Instead of exploring a space and questioning its rules, transformational creativity is about ignoring fundamental rules to come up with potentially impossible but highly creative ideas. Let’s keep on using transportation as an example. Instead of questioning the cost of air transportation, you may ask yourself: Why do cars have to park? Why do we need to travel at all? Transformational creativity has the potential to generate the most radical ideas.

In reality, we may very often be using a combination of these three types of creativity when brainstorming, and this is a good thing. By starting with transformational creativity, then moving onto exploratory and combinational creativity, you are not leaving any potential idea of the table, and can go from crazy to actionable creative ideas.

What about brainstorming as a team? The process is not too different, and may even be more powerful, since you’re combining the brain power of several people. It does need to be more structured, though. Make sure to have these safeguards in place:

• Create a safe environment. eEnsure everyone feels comfortable, and that there is no competition. Define acceptance as the default. Psychological safety is paramount for a productive and creative team. There are no bad ideas in a good brainstorming session.
• Avoid planting a solution. Don’t start with an example so people are not primed and coming up with similar solutions.
• Don’t shoot for the stars. “If you start a meeting and you say, ‘Okay, we are going to come up with really good ideas,’ that can be a really bad way to start,” says Christian Schunn of the University of Pittsburgh, who published an interesting paper about idea generation.

How to brainstorm in five steps

How does this all work in practice? Here is a step-by-step guide to effectively brainstorm and generate new ideas. Remember the principles laid out earlier: quantity versus quality, building a creative routine, and using all three creative modes to ensure you don’t leave any ideas off the table.

• Set your focus. Define the problem or area you will be looking at. It can be as narrow as a specific annoyance you face in your life, and as broad as a whole industry, but you can’t just have a vague brainstorm with no predefined focus.
• Gather new material. Give yourself — and the team if it’s a group brainstorm — time to familiarise yourself with the area of focus. This means reading articles, watching videos, etc. If it’s a group brainstorm, this step should ideally happen before the session to give time to your brain to incubate these ideas, but if not you can block a bit of time at the beginning of the session.
• Generate ideas. Remember, quantity over quality. Use the three creative modes presented earlier. Combinational to mix old ideas together, exploratory to investigate new potential ideas within the rules of a given space, transformational to break the rules and come up with radical ideas.
• Test your ideas. This is where most brainstorming sessions fail to take the one extra but necessary step. Instead of selecting your ideas on the spot, you need to test them in the real world. Select the few most promising candidates, and see how your audience reacts. For a book, write a blog post. For an app, build a landing page or a quick MVP.
• Select and refine your ideas. Use the feedback your receive to adapt or drop your ideas. If a particular problem or area keeps on coming back in the feedback… Go back to step 1.

Whatever you do, keep sharing your ideas with the world. Don’t fear that people will steal them. Since most ideas are combinational, chances are yours are not new, and only execution will matter. If you do come up with a truly transformational idea, it’s very unlikely that someone will be able to pick it up and just run with it. So go forth and multiply your ideas!

Join 100,000 mindful makers!

Ness Labs is a weekly newsletter with science-based insights on creativity, mindful productivity, better thinking and lifelong learning.

One email a week, no spam, ever. See our Privacy policy .

Don’t work more. Work mindfully.

Ness Labs provides content, coaching, courses and community to help makers put their minds at work. Apply evidence-based strategies to your daily life, discover the latest in neuroscience research, and connect with fellow mindful makers.

Ness Labs © 2022. All rights reserved .

10 Brainstorming Techniques for Developing New Ideas

Start collaborating with Mural today

Brainstorming is an essential practice for creative thinking and problem-solving. At its most basic, brainstorming simply means identifying a problem, and then coming up with as many fresh ideas as possible that may help solve that problem. In practice, however, brainstorming needs to be structured to be successful.

No matter which brainstorming method you try, be sure to keep these tips in mind:

• Outline a problem or topic that needs further examination and group feedback for your brainstorming session.
• Use a shared digital space to conduct your brainstorming . This not only makes brainstorming easier, but also ensures you don’t lose any ideas and have a document to refer to in the future.
• Encourage quantity over quality . Participants can often get hung up on having the “best idea” but this often leads to creative blocks. Instead, focus on generating a high number of ideas first. Once you have a lot of ideas, you can start refining the best ones. However, if you have fewer ideas, you’ll have fewer options to choose from.
• You don’t always need a facilitator, but having one can help to have somebody guide a group of people through the ideation process . If needed, assign a team leader or facilitator before the idea generation begins to keep track of time limits and record the most creative ideas.

10 brainstorming techniques for better idea generation

Question brainstorming.

Question brainstorming is a method where participants come up with as many questions about the problem as possible. In this activity, participants should only come up with questions, not answers. This method is great because it helps the group try to understand the problem completely without the pressure to have the single best solution.

How to try this technique:

• Decide on the central topic or problem you want to brainstorm for.
• Each group or individual then takes 10-15 minutes to write as many questions about the topic or problem they have without trying to come up with an answer to any of them.
• After 10-15 minutes, the group gets back together to share, group, and prioritize the questions.
• Go through the full list of questions and start answering each one to better understand the core topic or problem.

Need inspiration? Check out these 25 brainstorming questions for effective brainstorms .

Hybrid brainstorming

The best way to get the most out of your brainstorming efforts is to combine the best of async and real-time idea generation. This technique starts off with individual brainstorming and ends with a collaborative, group brainstorming session.

How to do it:

• Chose the top 5 questions you want the brainstorming session to be based around.
• Have each participant begin brainstorming in their own dedicated space.
• Next, have everyone share their ideas so you have one central repository of all the ideas.
• Go through each question and discuss any questions that arise.
• Vote on the top ideas to focus on.
• Turn the top ideas into action items and celebrate your successful ideation session.

Mind map brainstorming

Using the mind mapping brainstorming technique helps your team use the central focus of your project as a starting point, explore all the potential variables tied to that project, and build an understanding of how those variables relate to the main topic, as well as to one another.

• Add a central topic or question to the center of the mind map.
• Add branches and nodes based on secondary thoughts to the primary topic. 
• Keep adding branches and nodes based on additional ideas until you have a satisfactory number of related ideas.

Starbursting

The starbursting method involves asking a series of questions, typically using the 5 W's and 1 H (Who, What, Where, When, Why, and How) to explore the different aspects of the problem or topic. Starbursting is a divergent thinking approach focused on asking questions about the topic in order to generate a wide range of ideas.

How to run a starbursting exercise:

In Mural , Add six sticky notes around a central idea or problem and label them “Who,” “What,” “Where,” “When,” “Why,” and “How.” For each question, encourage participants to generate as many ideas as possible. Alternatively, you can conduct starbursting in Mural by creating a mind map.

Once the ideas have been generated, the facilitator can use other ideation techniques to further refine the ideas and identify potential solutions to the problem. Try clustering similar ideas together, categorizing the ideas into different groups with tags, or prioritizing the ideas based on their potential impact or feasibility.

The 'crazy eights' brainstorming technique is a great way to explore a wide range of ideas by encouraging quantity and time-boxing every exercise for efficiency. Brainstorm 8 different solutions in just 8 minutes. Once you’re ready, you can review the ideas and agree on the most effective solution.

How to do a crazy 8s brainstorm:

• Have each participant take a piece of paper (or use an online whiteboard ) and divide it into eight parts.
• Sketch or detail eight ideas, spending one minute for each panel.
• Have each participant share their own ideas.
• Vote on the top ideas and turn them into actionable next steps.

Round robin brainstorming

A round robin is a great way to encourage your team members to think critically about a problem or challenge, as well as build on one another's ideas. The format calls for dividing your team into small groups, and having each person pass their ideas along to the following teammate, who then offers counterpoints or further suggestions, helping identify potential weaknesses or spurring innovation within the session.

How to run a round robin:

• Set the brainstorming topic or problem you’re looking to solve.
• Divide your group or participants into four teams, or add more panels to accommodate more participants.
• Have each group or participant write down a proposal and move to the next panel
• Looking at what the other group or participant suggested, write down reasons why their proposal might fail then move to the next panel.
• Based on the proposal and weaknesses, have each participant or group craft a final concept using the weaknesses as feedback.

Looking for more instruction? Check out our complete guide on round robin brainstorming with a walkthrough, tips, and variations of the round robin method.

6-3-5 brainwriting

6-3-5 brainwriting is a group brainstorming method that involves creating a lot of ideas and building on the ideas of the other participants to give a total of 108 ideas in a short amount of time.

To run this exercise, you'll need 6 participants to create 3 ideas each within 5 minutes. Run this exercise for 6 rounds for a quick, half-hour brainstorming exercise.

• Have each participant select one panel and begin writing ideas related to the topic or problem statement on the first row of sticky notes.
• After five minutes, participants move to a different panel and do another round of ideation, using the previous row of ideas for inspiration.
• Run four more rounds of this exercise, or stop when you have enough ideas.
• Cluster and vote on the winning ideas with your team.

Brainwriting can be a marked improvment over more conventional brainstorming methods. In addition to reducing the pressure to openly share ideas that might be a bit "out there," brainwriting can help reduce participation inequity and prevent groupthink. Learn more in Mural's guide to brainwriting .

Reverse brainstorming

Reverse brainstorming involves looking at a problem from a different angle for generating ideas. Instead of asking, "What can we do to solve this problem?" the goal becomes, "How can we create this problem or make it worse?"

By focusing on the opposite of the problem or goal, team members can uncover new insights and approaches that they may not have considered otherwise. Once the "reverse" ideas have been generated, participants can then work to flip them around and find ways to turn them into positive solutions.

How to run a reverse brainstorm

• Draft a problem statement or starting topic for your team to brainstorm on
• Instead of posing the question: “How can we fix this?” ask “How can we make this problem worse?
• Participants will instead find more root causes driving the problem that you can start to address and dig deeper into
• Review and prioritize the ideas by severity or impact

Silent circuit

The silent circuit is a quiet brainstorming activity helps groups ideate across multiple topics while still being inclusive for quieter participants. This method is great for large groups, hybrid teams, and teams with introverts. This method is also great for getting different points of view.

How to run a silent circuit:

• Write "how might we" questions or a different prompt at the top of each brainstorming section.
• Set a timer and encourage participants to pan around the canvas and silently add as many ideas on sticky notes as they can under each prompt.
• When time is up, participants return to their original question and share all the ideas for each category.
• Review the questions and create action items for the best ideas.

Brain-netting

Brain-netting is a traditional brainstorming session conducted online. Online brainstorming isn’t a new concept, but recent shifts in remote work and collaboration tools have made this a much more common exercise. The main benefit of a brain-net is that it can be done asynchronously, meaning collaborators don’t have to. 

Brain-netting is more common among remote teams, where asynchronous communication and video calls are the norm. However, any team can leverage brain-netting to take a brainstorming activity online, generate innovative ideas, and get the creative juices flowing.

How to run a brain-netting exercise:

• Kickoff the session and warm-up your team with a virtual icebreaker that can be done asynchronously
• Add a problem statement or define a topic to ideate over
• Brainstorm together or set a deadline for when team members should have added their ideas to the brainstorming platform 
• Synthesize the ideas together, or summarize them and share the common themes
• Prioritize the ideas, pick a winner, and/or define next steps

Related: How to Facilitate a Brainstorming Session

Why brainstorming is essential for innovation

Brainstorming fosters creative thinking.

If you’re facing a difficult problem, brainstorming can help you to generate potential solutions that you might not have thought of otherwise. The free-flowing nature of brainstorming is meant to encourage exploration and a diversity of ideas — even those suggestions that seem tangential or unrelated at first may wind up forming the basis for effective solutions later on, or as inspiration for new products or features.

Brainstorming puts your best ideas in one place

Brainstorming helps you to organize your team's thoughts and feedback on any project. By structuring your brainstorming sessions so that everyone is engaged and all ideas are recorded, you can later organize your feedback by theme , using tools like tags for sticky notes and filtering. This can help you to better assess which ideas are worth pursuing and which ones are not, and begin to quickly and easily outline actionable next steps.

Types of brainstorming

There are three main types of brainstorming. Each approach has advantages and disadvantages, though hybrid brainstorming (enabled by platforms like Mural) mitigates the issues of by synchronous and async brainstorming.

Asynchronous brainstorming

Individuals are presented with a question or topic to consider on their own. Later, team members contribute their ideas to a shared document for further discussion and organization. The main advantage with asynchronous brainstorming is that it inherently avoids issues like groupthink since each individual completes the exercise independently.

Synchronous brainstorming

A team gathers together to brainstorm in a meeting, with everyone sharing ideas and grouping suggestions by theme. The main advantage of synchronous brainstorming is that it allows your team to build on one another's ideas in real time, making your session potentially more productive and speeding up innovation.

A hybrid approach to brainstorming combines elements of both asynchronous and synchronous ideation. By establishing a basis of psychological safety, as well as taking advantage of modern, cloud-based collaboration tools and features built to avoid groupthink, it's really possible to get the best of both worlds.

Tips for productive brainstorming

Establish an environment of psychological safety.

One of the most important elements in unlocking effective brainstorming is establishing a culture and environment of psychological safety . A brainstorming session should be one where everyone feels comfortable contributing without self-editing — in this phase, ideation shouldn't be interrupted by critiques or pushback. It's simply about collecting as many ideas and different perspectives as possible.

Psychological safety also means you are less likely to be impeded by groupthink — if the brainstorming session is characterized by a wide array of ideas and even constructive disagreement, you will have a much better basis upon which to formulate potential solutions than if everyone agrees or one line of thinking dominates your discussion.

Use private brainstorming within the context of a group meeting

To achieve this, you'll need to make people feel safe to share what they may think are outlandish or controversial ideas, which can be difficult to achieve. If you're not sure how best to begin, tools like Private Mode when using Mural can help you avoid groupthink by hiding the feedback that others are providing. Alternatively, when using a shared digital space like Mural, asynchronous brainstorming is also a viable solution — this allows people to reflect on their own, and bring their unique perspectives without outside influence into the meeting at a later date for discussion and organization.

Related: 7 Key Rules for Brainstorming

Follow-up after the brainstorming session

Brainstorming ideas is only the first step. After the brainstorming session is over, the team moves into the implementation phase. Be sure to define next steps and the roles of each team member so everyone understands the workflow and what’s expected of them. By following up, you ensure that the ideas you captured don’t get forgotten and a solution can be implemented.

Define a strategic goal for your brainstorming meeting

Once you've established the approach for your brainstorm, it helps to give greater context to your ideation by defining an overarching, strategic goal. Are you at the very outset of a problem with a lot of variables, and trying to better understand how they relate to one another? Is the purpose of your meeting to discover new ways to improve user experience for a given product or feature? Each use case requires a different basic framework for your brainstorming meeting.

Use these techniques to run better brainstorming sessions

Brainstorming is an essential part of the innovation process, but it can be difficult to come up with new ideas if you’re not sure where to start. The techniques we’ve outlined in this post should help you structure your brainstorming sessions in a way that makes them more effective and helps you produce actionable insights and takeaways.

If you want to make your brainstorming sessions even more productive, using a shared digital space like Mural not only unlocks visual thinking and online collaboration, but also builds in strategies to combat issues with groupthink and allows for hybrid brainstorming sessions that combine the best of asynchronous and real-time meetings.

Mural offers a host of brainstorming templates to kickstart more effective and meaningful (not to mention fun) sessions. Get started today with a Free Forever plan , and invite unlimited guests so you can build the next great idea together with your whole team.

Related blog posts

7 ground rules for brainstorming

How to Facilitate a Brainstorming Session

How to encourage thinking outside the box

Related blog posts.

How to Collect Feedback from Stakeholders Effectively

Visual collaboration: What it is & how to get started

What is the difference between traditional project management and Agile?

Ultimate Guide to Boosting Your Creativity To Generate Great Research Ideas!

Introduction.

Research is a fundamental aspect of academic and scientific pursuits. It involves the systematic investigation of a particular subject or problem to discover new knowledge or validate existing theories. However, the success of any research project heavily relies on the quality and novelty of the research ideas generated. Coming up with innovative and compelling research ideas can be a challenging task for researchers and scholars. It requires a combination of creativity, critical thinking, and a deep understanding of the research domain. Fortunately, there are various techniques and tools available to assist in the generation of research ideas.

In this article, we will explore the concept of research idea generation and its significance in the research process. We will discuss the challenges researchers face in generating ideas and the different techniques they can employ to overcome these challenges. Additionally, we will delve into the world of AI-powered research idea generators and how they can enhance the idea generation process.

Why Research Ideas Are Important?

Research ideas are the foundation of any successful research project. They provide the direction and purpose for the study, guiding researchers in their quest for knowledge and understanding.

Identifying Research Problems

One of the key reasons why research ideas are important is that they help to identify research problems. A research problem is the specific issue, contradiction, or research gap that the study aims to address. It gives the research a clear purpose and justification. Research ideas also play a crucial role in shaping the research questions . Research questions are specific and actionable inquiries that guide the research process. They help researchers focus their efforts and ensure that the study is relevant and meaningful.

Shaping Impact and Contributions

Furthermore, research ideas contribute to the overall significance of the research. They help researchers identify the potential impact and benefits of their study. By generating innovative and valuable research ideas, researchers can make meaningful contributions to their field of study and society as a whole.

Stimulating Creativity And Innovation

In addition, research ideas stimulate creativity and innovation. They encourage researchers to think critically, explore new perspectives, and develop novel approaches to solving problems. Research ideas push the boundaries of knowledge and drive scientific progress.

Securing Funding For Research Projects

Moreover, research ideas are essential for attracting funding and support. Funding agencies and organizations are more likely to invest in research projects that have well-defined and compelling ideas. Research ideas that address pressing societal issues or have practical applications are particularly attractive to funders.

Challenges in Generating Research Ideas

Lack Of Inspiration Or Creativity

Generating research ideas can be a challenging task for many individuals. One of the main challenges is the lack of inspiration or creativity. Coming up with unique and innovative ideas requires thinking outside the box and exploring new perspectives. However, this can be difficult when faced with limited knowledge or exposure to different fields of study.

Fear Of Failure Or Rejection

Another challenge in generating research ideas is the fear of failure or rejection. Many individuals hesitate to explore unconventional or risky topics due to the fear of not being able to find enough information or support for their ideas. This fear can hinder the generation of creative and groundbreaking research ideas.

Overwhelming Amount Of Information Available

Additionally, the overwhelming amount of information available can also pose a challenge in generating research ideas. With the vast amount of literature and research already conducted, it can be daunting to find a gap or niche that has not been explored yet. This can lead to a sense of discouragement and difficulty in generating unique research ideas.

Time Constraints And Limited Resources

Moreover, time constraints and limited resources can also be obstacles in generating research ideas. Research projects require time, effort, and resources to be conducted effectively. Limited access to funding, equipment, or research materials can limit the scope of research ideas that can be pursued.

Lack Of Guidance Or Mentorship

Lastly, the lack of guidance or mentorship can also hinder the generation of research ideas. Without proper guidance or mentorship, individuals may struggle to identify their research interests or find the necessary support to develop their ideas. This can result in a lack of motivation and direction in generating research ideas.

Exploring Different Research Idea Generation Techniques

When it comes to generating research ideas, there are various techniques that can be employed to stimulate creativity and uncover unique topics. In this section, we will explore some of the most effective research idea generation techniques.

Brainstorming

Brainstorming is a popular technique that involves generating a large number of ideas in a short period of time. It encourages free thinking and allows for the exploration of different perspectives and possibilities. During a brainstorming session, participants are encouraged to share their ideas without judgment or criticism.

Mind Mapping

Mind mapping is a visual technique that helps to organize thoughts and ideas. It involves creating a diagram that branches out from a central idea, with each branch representing a different aspect or subtopic. Mind mapping allows for the exploration of connections between ideas and can help to generate new insights.

SCAMPER is an acronym that stands for Substitute, Combine, Adapt, Modify, Put to another use, Eliminate, and Reverse. This technique encourages researchers to think creatively by asking questions and exploring different possibilities. By applying each of the SCAMPER techniques to a research topic, new ideas and perspectives can be generated.

Random Word Association

Random word association involves selecting a random word and then brainstorming ideas that are associated with that word. This technique helps to break free from conventional thinking and can lead to unexpected and innovative research ideas.

SWOT Analysis

SWOT analysis is a technique commonly used in business and marketing, but it can also be applied to research idea generation. SWOT stands for Strengths, Weaknesses, Opportunities, and Threats. By analyzing these four aspects, researchers can identify potential research topics and areas of interest.

Reverse Brainstorming

Reverse brainstorming is a technique that involves identifying potential problems or obstacles related to a research topic, and then brainstorming ideas to solve or overcome those problems. This technique can help researchers to think critically and uncover unique research ideas.

Role Playing

Role playing involves assuming different roles or perspectives related to a research topic. By putting themselves in the shoes of different stakeholders or individuals, researchers can gain new insights and generate fresh ideas. Role playing can be done individually or in a group setting.

Delphi Method

The Delphi method is a structured technique that involves gathering input from a panel of experts through a series of questionnaires or surveys. The responses are then analyzed and summarized to generate research ideas. This technique helps to ensure a diverse range of perspectives and can lead to well-informed research topics.

Utilizing AI-Powered Research Idea Generators

In today’s digital age, artificial intelligence (AI) has become an integral part of various industries, including research. AI-powered research idea generators are tools that leverage the power of AI algorithms to generate unique and innovative research ideas. These generators analyze vast amounts of data, including research papers, articles, and other relevant sources, to identify patterns, trends, and gaps in existing knowledge.

By doing so, they can provide researchers with a wide range of research topics and ideas that they may not have considered otherwise. This capability aligns seamlessly with the principles of the Fourth Industrial Revolution (IR) 4.0 , which emphasizes the integration of digital technologies and intelligent automation in various sectors, including academic research. This integration facilitates a more efficient, innovative, and data-driven approach to discovering new research opportunities.]

Examples of AI-Powered Research Idea Generators

Developed by OpenAI, Chatgpt is a conversational model that can help brainstorm and refine research topics based on the input you provide, simulating a discussion to expand on ideas.

Gemini is an AI-powered tool that is designed to generate innovative research ideas by analyzing existing literature to identify trends, gaps, and correlations, helping researchers find unique angles and untapped areas of study.

Appy Pie’s free AI research title generator offers the capability to generate over 500 tailored research title ideas, making it easy to start your project with a compelling title.

Advantages Of Using AI-Powered Research Idea Generator

Saves Time And Effort in Idea Generation

One of the key advantages of using AI-powered research idea generators is their ability to save time and effort. Instead of spending hours brainstorming and searching for research ideas, researchers can simply input their preferences and requirements into the generator, and it will generate a list of relevant and interesting research topics.

Overcoming The Struggle To Come Up With New Research Ideas

Furthermore, AI-powered research idea generators can also help researchers overcome the challenge of writer’s block. Sometimes, researchers may struggle to come up with new and innovative ideas for their research projects. AI-powered generators can provide them with fresh perspectives and insights, sparking their creativity and helping them overcome any mental blocks.

Diversity in Research Ideas

Another benefit of utilizing AI-powered research idea generators is the diversity of ideas they can generate. These generators have access to a vast amount of data from various disciplines and fields of study. As a result, they can generate research ideas that span across different domains, allowing researchers to explore interdisciplinary topics and make connections between seemingly unrelated fields.

Bear In Mind: AI Is A Tool, Not a Crutch

However, it is important to note that while AI-powered research idea generators can be a valuable tool for researchers, they should not be solely relied upon. These generators are meant to assist researchers in the idea generation process, but the final decision on the research topic should still be made by the researcher based on their expertise and interests.

In conclusion, enhancing your creativity is pivotal in generating great research ideas that can transform ordinary inquiries into groundbreaking studies. By embracing diverse thinking styles, leveraging technology like AI-powered idea generators, and continuously seeking different perspectives, researchers can cultivate a fertile ground for innovation.

Remember, the key to unlocking the potential of your research lies in the boldness to explore and the dedication to persistently challenge conventional boundaries. The journey to create significant, insightful, and impactful research ideas is a continuous process that evolves over time, applicable to both newcomers and seasoned researchers alike. Embrace this journey with curiosity and an open mind, as it can lead to discoveries that potentially reshape your field of study.

Leave a Comment Cancel reply

Save my name, email, and website in this browser for the next time I comment.

Related articles

How to Formulate Research Questions in a Research Proposal? Discover The No. 1 Easiest Template Here!

7 Easy Step-By-Step Guide of Using ChatGPT: The Best Literature Review Generator for Time-Saving Academic Research

Writing Engaging Introduction in Research Papers : 7 Tips and Tricks!

Understanding Comparative Frameworks: Their Importance, Components, Examples and 8 Best Practices

Revolutionizing Effective Thesis Writing for PhD Students Using Artificial Intelligence!

3 Types of Interviews in Qualitative Research: An Essential Research Instrument and Handy Tips to Conduct Them

Highlight Abstracts: An Ultimate Guide For Researchers!

Crafting Critical Abstracts: 11 Expert Strategies for Summarizing Research

How to Generate Ideas That Work

• Click to share on LinkedIn (Opens in new window)
• Click to share on Facebook (Opens in new window)
• Click to share on X (Opens in new window)
• Click to email a link to a friend (Opens in new window)
• Click to print (Opens in new window)

Babson professor Heidi Neck wants to help budding entrepreneurs learn how to identify and develop high-quality ideas. Her solution? IDEATE—the jumping-off place for how to generate ideas of value.

“I often hear that execution is what separates entrepreneurs from non-entrepreneurs. I don’t disagree, but I also see a lot of failures among those that do start the business. Why so many failures? It’s because the idea was bad to start with!” said Neck.

In their new book, The IDEATE Method , authors Dan Cohen, Greg Pool, and Neck provide a proven road map for how to generate ideas that can be replicated and taught.

From Ideation to IDEATE

Generating new ideas that create value is at the core of entrepreneurship. And ideation is at the core of new ideas. Yet many are at a loss when it comes to knowing how to generate ideas.

“Experienced entrepreneurs generate valuable ideas. They develop skills. They have certain innate abilities related to pattern recognition. They see a pattern—an opportunity—where others see chaos or simply miss the opportunity. They have domain expertise, a professional network, and industry experience that students lack,” said Cohen, a professor at Wake Forest University. “With the IDEATE method, students can spot, evaluate, and select ideas that they would otherwise miss.”

At Wake Forest University, Cohen noticed that his students had lots of success with a teaching approach focused on ideation. They knew how to generate ideas and they knew how to develop those ideas.

Cohen decided to formalize his approach. He wanted his ideation model taught in other settings. Working closely with entrepreneur and educator Greg Pool, and then Neck, the IDEATE Method was born. The secret to ideation was unlocked for all to use. The question of how to generate ideas became something teachable.

Ideation for All

Neck believes that being able to spot problems, knowing how to generate new ideas, and evaluating their value is a skill set that requires practice.

The IDEATE Method helps students do just that. It’s an ideation system. It starts with problem recognition and uses concepts of deliberate practice and repetition to help students generate high-potential ideas. It goes way beyond brainstorming, instead giving students a toolkit for how to generate ideas.

Without ideation, it’s difficult to know whether an idea has potential. It’s tempting to lean toward novelty. With ideation, people can focus on profit potential and market size. It’s an approach that has been validated through research. And, it’s been found to be significantly more effective than other ideation methods because it takes the guesswork out of how to generate ideas.

What Is the IDEATE Method?

In its most simple form, IDEATE is a structured methodology most useful for novice ideators and entrepreneurs—those who have less experience. These are people who need to know how to generate ideas. The book shock-tests 100 ideas to assess their viability and market potential. With self-directed exercises, reflection, and evaluation worksheets, students not only develop their skill set in opportunity generation, but also build their entrepreneurial mindset.

Ideation Action Steps

• Identify – Distinguish between high-quality and low-quality problems
• Discover – Actively search for ideas in problem-rich environments
• Enhance – Add innovation and novelty to enhance existing opportunities
• Anticipate – Use the four sources of change to anticipate new opportunities
• Target – Identify target customers and understand their unmet needs
• Evaluate – Practice scoring, selecting, and defending high-quality ideas

Industry experience is important, as is a strong network. But most aspiring entrepreneurs, especially young students in entrepreneurship courses, don’t have those attributes yet so they need to acquire new tools. Neck’s ideation workbook provides a proven method for how to generate ideas that can produce real value. She will be unveiling the IDEATE methodology at upcoming Babson Academy programs this year.

Posted in Insights

Tagged Entrepreneurship Education , Faculty

More from  »

Related Stories

Latest stories.

A Game-Changing Gift Seeks to Take Babson Golf to the Next Level

All About the Students: A New Role for a Campus Leader 

I Won $100,000 for My Business Pitch—and 3 Simple Strategies Helped Me Do It

More babson.

Stack Exchange Network

Stack Exchange network consists of 183 Q&A communities including Stack Overflow , the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.

Q&A for work

Connect and share knowledge within a single location that is structured and easy to search.

How to come up with research ideas?

As a very new researcher who is exploring the best way to generate ideas, some guidance on this question would be very helpful. I have found that this is NOT easy. Ideas seem to pop out of my Professor every day and I wonder how he does it. This question is broad;

How do you tend to come up with initial/seed ideas? What is your search method (if you have one)?

What proportion of your ideas for past papers come from; (i) colleagues, (ii) intentionally browsing the literature for ideas, (iii) on the spot inspiration, (iv) conferences, (v) other?

How do you prioritize research ideas?

Is there any special, generalizable method that you've discovered to sift out those ideas that are likely to be unrealistic early on in the process of idea generation?

Based on small amounts of anecdotal evidence I have reason to believe that there is vast heterogeneity among professors regarding the above questions. For example, economist Steven Levitt says he works on 22 papers at once. A professor I know will have maybe 25% of this at any one time.

Related but not duplicate: Is there any software or tools for managing developing research ideas?

• research-topic

• 11 Anecdotally - for me it just happens (and I have much more ideas than time to develop them), while either reading or (much more) solving other problems. Just sparks of "what if?" or "can I generalized it?". For me it rarely happens on purpose - it it not hard, but impossible to force myself to be creative (on research or anything else). Related - Paul Graham, "How to get startup ideas" . –  Piotr Migdal Commented Dec 18, 2012 at 18:53

9 Answers 9

Okay, as you say, this is very broad, and possibly argumentative. So, I'll try to section off my answer for your various sub-questions, and talk not so much about how I do come up (and organize) research ideas, but how I see it done by everyone (including me).

Coming up with ideas

The most exciting phrase to hear in science, the one that heralds new discoveries, is not “Eureka!” but rather, “hmm... that's funny...” — Isaac Asimov

It's probably very akin to asking a large number of artists “how do you come up with inspiration?” , i.e. you can probably get one thousand different answers, and yet not useful answer at the same time. However, there are some elements that I think are common to all. You can't “trigger” new ideas to come into your mind, but you can put your mind into the right disposition to host these new ideas: recognize them and welcome them. Below is a list, certainly partial and limited, trying to detail my perspective in this matter:

Be challenged! Nothing sparks ideas more than being confronted with contradiction, healthy criticism, a spirited debate, maybe a bit of competition. Some people manage to do that by themselves, arguing against their ideas and improving them. I myself (and most of the colleagues and students I have seen) need an echo chamber, someone to discuss things with. If they're not exactly from your field, all the better, as they may have unusual/naïve/silly questions or expectations.

To give an example, some of the most “successful” ideas I have had came while answering questions, for example from a PhD student or colleague, and replying by “no, it doesn't work like that… in fact, it's probably always guaranteed to be false, because… see, it's linked to X… or maybe it's not? hum…”

Be curious! Ideas come from problems. Identifying worthy problems in your field of research, and dissecting larger issues into of specific problems of manageable scope, is at least as hard as coming up with new ideas. In the end my feeling is that, especially for a researcher, all ideas are the result of one’s curiosity.

Manage to get some free time for thinking (and not: teaching, supervising, tutoring, reviewing, writing, sleeping, …). Body and mind. Sure, an idea can pop into your head any time, but it's probably less likely to happen when you teach basic calculus all day that when you get some time to really think .

Know your field, know where a new development need to occur, what is currently missing. Read review papers, search for such ideas through people's articles or blog posts , discuss with senior colleagues who have a comprehensive view of the field, …

One of the ways you can come with ideas is by analyzing how different groups work in your fields, seeing what has been addressed and avoided, what big questions are still open, and how you can link between different works to build a coherent global picture… This is not always successful, but it usually generates some good ideas along the way!

Explore more or less closely related fields, and see if there is something from your background that you could apply to their problems, or ways you could build something together. Such ideas tend to be very strong, because you can oftentimes apply an entire branch of knowledge (ideas, methods, algorithms, etc.) to a very different problem. In that case, the added value comes from your different perspective, as you might try things that others would not think of.

Ways have been devised to come up with new ideas on a given topic, either alone or in group sessions. Brainstorming is probably the best know such method (and might be the most popular, in one form or another), but a really large number of creativity techniques have been developed. They can be applied both to enhance creativity or to boost problem solving efficiency.

Organizing ideas

A quote often attributed to Kant: “someone’s intelligence can be measured by the quantity of uncertainties that he can bear” . If that true, that has serious consequences for research. Accepting that your mind can only efficiently support a finite number of ongoing research ideas, you have to come up with ways to write them down, organize them, prioritize them, come back to them later, etc. Just as you cannot juggle with as many balls as you'd like, such “external” tools will help your brain focus on the ones that you assign high priority (or the ones to which it gives high priority; the brain works in funny ways).

Most people use very low-tech tools for that:

Notebooks , either sorted chronologically or thematically; in the later case, open a series of blanks pages for each new project/idea, and flip through the book whenever you want to check on them. I use a Moleskine ( WP ) for that purpose; having a nice, leather-bound notebook somehow helps me “value” it more and treat it with care (always have it with me, actually use it).

Post-it’s scattered through one’s (real or virtual) desktop. Downsides are obvious.

More people than I thought actually don't use any tools, and just keep all in their mind. Apparently it can be done, but I don't advise it.

But more complicated methodologies have been devised, that are supposed to help you with it:

• Mind mapping , either on paper or software-based.
• Using todo-list flat or two-dimensional todo-list software, or more complex task-tracking software (see, e.g. Trello ).
• The software side of this question is already covered (though possibly not extensively) here on this very Q&A site .

Finally, don't underestimate the possibilities opened by delegating: people in charge of a specific project or sub-project (PhD students or post-docs) can be tasked with maintaining a list of ideas by all contributors of the project, to come to later on.

Answers to your miscellaneous smaller questions:

Most ideas are hardly “traceable” to one source or another. A given idea might have formed in my head during a conference, seeing how people were failing to address a certain issue, then crystallized during a discussion with colleagues, but would never have occurred to me if not for a literature review I had performed a few months before.

I'll come back a bit later and continue working on this answer :)

• 5 +1 for the Asimov quotation. I had this as an epigram in my PhD thesis. –  Nicholas Commented Dec 19, 2012 at 9:37
• Agreed! +1 for Asimov! –  Ben Norris Commented Dec 19, 2012 at 11:44
• What do you think of using something like OneNote as a surrogate to the physical notebooks you were recommending? Latex equation typesetting websites can effortlessly generate .gif pictures of your equations that you can copy into OneNote. Wouldn't this do the same thing with the same level of efficiency (or even more because you won't have any clutter) as a physical notebook? –  Jase Commented Dec 23, 2012 at 3:26
• I've only been using it for a few days, but TiddlyWiki seems promising as a non-linear notebook for ideas. –  Detached Laconian Commented Jul 21, 2018 at 7:07

I'll address two points in your question (the overall question is quite broad):

Ideas seem to pop out of my Professor every day : If you've worked on enough problems, you amass a collection of tools and mental shorthands that you can apply to a new problem. It's a matter of experience. You also might see someone else's paper and realize that they are doing something in a clumsy way and you have learnt a better way to do it, and so on.

I wouldn't worry too much about this: it's a matter of time and experience, and will happen on its own. You're not evaluated on the number of ideas you have in any case. You might want to check how many of these ideas are actually good ones :).

How do you tend to come up with initial/seed ideas? : When you're first staring at a problem, it can be intimidating and difficult. While there's no single strategy for getting a "leg up", some useful techniques (and these might be very math/CS specific) are:

• simplify the problem : can you solve a simpler version ? if not, can you simplify even further ? Often, finding the largest solvable element starts to get your mind rolling
• pattern match : does this problem look like something related that has been solved ? can you borrow a method from there ? if not, why not ? again, the goal is to get your mind off the "ZOMG THIS PROBLEM IS TEH HARD" and onto "Here's a tiny piece that I can chew on".

I'm sure others will have useful ideas as well. Ultimately, you'll find that getting ideas isn't the problem: it's getting GOOD ideas that is hard.

• 1 +1 for pattern match; found it very effective specially among different graph-like representations. –  seteropere Commented Dec 20, 2012 at 7:15
• +100 because I expect that your simplify the problem and pattern match advice will be very helpful. –  Jase Commented Dec 24, 2012 at 5:04
• Check out e.g. Pólya's "How to solve it" for related suggestions –  vonbrand Commented Jan 11, 2016 at 0:10

Here is some things I found useful:

Attending public seminars at the department could spark nice ideas (even if it seems not related to your research).

Chatting with other graduate students.

• Reading deeply with why? in mind. This means reading a lot and also means stopping more than usual in the assumptions hypothesis and results for different papers.
• Read future work and conclusions of the papers. Some papers have a real list of future research ideas.
• Capture the Big Picture. This usually will result in many whys for what you encountered.
• Ask Questions .. Even what seems as silly and fundamental questions for some can be the key for good ideas.

Do you, when presented with anything related to your research, routinely ask, "Why is that? How can I tell if that is the case?"

If not, try doing so.

If so, you probably won't be short on ideas. Your problem will be sorting the good ones from the bad ones.

Then start asking "Is this important? How can I explain why it is important?"

Be open/curious to what related disciplines are doing. In some you'll see that the way they solve their problems could also applied to your field but hasn't been tried yet.

I am also a green researcher, and similarly to you, I find coming up with ideas a daunting task. I have tried to approach this task in a bit more systematic way, than to just be waiting for Godot. Feel free to draw inspiration!

• Finding a problem to work on

I keep a list of interesting problems. This could be something a hear about at a seminar, read about in an article, or just something I think about. I write it down - usually half a page, only few references - and forget about it. I can then pull out my list, and find something. Some of the problems quickly turn out to be too small to be interesting, others not.

• Starting out

When one of my problems are deemed interesting enough, I turn to lit. study. This goes on until I find someone with an interesting treatise. Then I read it, and try to reproduce the result as they do it. (I should here mention that my field is theoretical)

• Reproducing - talk about it

I can use quite some time reproducing previous authors' work. But it is very fruitful, and you tend to learn something. I try to give a local seminar about the work at this point, the junior people in my department does bi-weekly blackboard seminars, where presentations like this are encouraged for exactly this reason.

At this point it is hopefully possible where I can go in and improve state of matter. So I start. This can sometimes require correspondence with the author of aforementioned work.

• Talk about it - again!

For me, discussing my work with peers is essential. At this point I would try to sneak in two slides about 'ongoing work' in a conference presentation in order to get feedback from peers and seniors from the field.

From this point on it is not so much getting the idea anymore, as following through on it. I will leave that to another day.

The most practical way is to go to the Library and look through journals for articles that interest you.

When you have found a selection, then sort them by a) Is this a current concern in your field? b) Is the prevailing methodology/technique practical - have you the resources? c) Will your supervisor(s) find this project interesting?

When you can answer all three questions as Yes, then do a deeper literature research and assess again whether the project is doable in the time available and publishable (sound and interesting to people in your field).

• 2 I think it would be much more efficient to do it over the internet where every single journal article can be accessed instantly (which is not the case in the library where some references are inaccessible or take 5 minutes of wasted time to find). –  Jase Commented Dec 30, 2012 at 1:21

Maybe you could try approaching the problem from another direction,

"What is it that you would like to achieve? what is the purpose of your research?"

There are millions of problems in life at the moment, and finding things to research is not the problem at all, even though it may seem that way. Inspiration is not purely found in a textbook, but are a function of the mind and soul and body.

Experience is what probably allows your professor to come up with constant questions. He probably practices free thinking, whereby he doesn't feel constrained in any way by other people and current belief systems. Maybe a lot of the problems that actually need to be understood, such as mental health and problems that people and our planet, experience everyday, just aren't being taken into your current world-view.

Science in itself is not an end. Science is a state of being, including understanding; and is a way that you as an intelligent, caring and investigative person (I presume) approach problems. A classic example of a problem is, that we don't understand. However, simply not understanding something is not a problem. A problem is something that has effects in the real world, such as, how can we help infertile couples reproduce and have children? Although it appears that now that we have resolved this in some detail, that it was the problem of not understanding DNA and the details of reproduction that probably is what resulted, with in vitro fertilisation, and even in vivo transplants etc. If one were to take the time to step out of this 'curiosity breeds progress' mindset, it would appear that these problems weren't purely driven by a quest for knowledge, but from real world problems, that have fortunately been solved.

I'd be interested in further discussion, as I have only this evening come up with an idea myself!

There's always a thirst for improvement, and this won't cease until people realize that happiness doesn't come from materials. Happiness is within all of us, all we have to do is tap into it. Being only 24 I have seen some truly eye-opening things and I am very humble to each of our personal strengths, but I do feel its a shame that research has become so fascinated with one-upmanship, and away from the real potential and benefit of being so intelligent.

I would recommend Pragmatic Thinking and Learning: Refactor Your Wetware by Andy Hunt. This book has some very good tips for coming up with ideas, research or otherwise.

• 5 I don't think this is very useful without giving some indication of what is actually in the book. –  David Z Commented Dec 21, 2012 at 2:09

You must log in to answer this question.

Not the answer you're looking for browse other questions tagged research-topic ..

• Featured on Meta
• Site maintenance - Mon, Sept 16 2024, 21:00 UTC to Tue, Sept 17 2024, 2:00...
• User activation: Learnings and opportunities
• Join Stack Overflow’s CEO and me for the first Stack IRL Community Event in...

Hot Network Questions

• Offline autocue prog for Windows?
• Why is the Liar a problem?
• Solaris 11 cbe: no more updates?
• Can RP2040 GPIOs be used as a "togglable ground" for NMOS Array?
• Do black holes convert 100% of their mass into energy via Hawking radiation?
• Do carbon fiber wings need a wing spar?
• How can I make the curves react to the texture's values?
• Browse a web page through SSH? (Need to access router web interface remotely, but only have SSH access to a different device on LAN)
• Are these colored sets closed under multiplication?
• I have two last names and want both Arxiv and google scholar to show them both everytime. How can I do this?
• ASCII 2D landscape
• crontab schedule on Alpine Linux runs on days it's not supposed to run on
• Fast leap year check
• How was Adam given the 7 Noahide laws without animals or women in existence?
• View undo history of Windows Explorer on Win11
• Little spikes on mains AC
• Arduino Uno Serial.write() how many bits are actually transmitted at once by UART and effect of baudrate on other interrupts
• Is it true that before European modernity, there were no "nations"?
• Smallest root of a degree 3 polynomial
• Are positive definite linear operator always invertible?
• security concerns of executing mariadb-dump with password over ssh
• Is Entropy time-symmetric?
• Which cartoon episode has Green Lantern hitting Superman with a tennis racket and sending him flying?
• How would you address the premises of Schellenberg's non-resistant divine hiddenness argument?

What is Idea Generation? – Definition, Techniques and Success Factors

Ideas are the key to innovation . Without them, there isn't much to execute and because execution is the key to learning, new ideas are necessary for making any kind of improvement.

It's obvious that ideas alone won't make innovation happen, as you need to be able to build a systematic process for managing those ideas. The point of ideation isn't just about generating tons of them but about paying  attention to the quality of those as well.

We agree that it can sometimes be difficult to come up with more of those high-quality ideas. It’s not unusual to get stuck in our old habits and routines when we’re supposed to be creating something new.

To break out of that rut, we've decided to look into idea generation as a whole as well as to introduce some of our favorite tips, tools and techniques that can be used to generate new ideas more systematically.

Table of contents

What is idea generation and why it's important, idea challenge, scamper technique, opposite thinking, brainstorm cards, analogy thinking, idea management tool.

• Define the problem or opportunity
• Involve the right people
• Set constraints
• Get the Idea Generation Toolkit

Idea generation is described as the process of creating, developing and communicating abstract, concrete or visual ideas.

As mentioned, ideas are the first step towards making improvement. Us making progress as individual  human beings   depends on new ideas.  From the perspective of an individual, new ideas can help you to move forward if you feel stuck with a task or are unable to solve a certain problem.

Perhaps you're looking for new ways to overcome a creative block or you're after a better solution to your problem? 

Maybe you need new ideas so that you can fully explore a new opportunity?

When it comes to your  organization , generating and collecting new ideas from your employees is the best way to uncover creative, tacit knowledge. The ability to create and develop new ideas allows you to:

• Stay relevant
• Make positive change happen in your organization

“Creative thinking inspires ideas. Ideas inspire change” – Barbara Januszkiewicz

Perhaps your organizational goal is to improve your and your team's efficiency or you need new ideas for making your product better?

Regardless of your goals or the types of ideas you're looking for, the purpose of new ideas is to improve the way you operate .

On a larger scale,   economies   depend on  innovation  to drive growth and increase well-being. Innovation creates new technologies and businesses, which provide new jobs for people.

So, although innovation isn’t about ideas alone, they are an important part of the equation as there wouldn’t be one without the other.

Tools and techniques for generating ideas

Okay, now you need ideas. What do you do?

The odds are that you’d run a brainstorming session. It has, however, been shown  that brainstorming not only takes more time and leads to less ideas, but also worse ideas than if the same participants had just tried to come up with ideas by themselves.

There are several other reasons why brainstorming may not be the best way to come up with ideas. Scheduling, organizing and documenting the session in a usable format will all take up even more time.

There are, however, some other useful techniques that might be worth giving a shot to challenge conventional thinking. Because you're going to need different ideas, it's good to have a few techniques in mind for generating them. Most of these ideation methods can be used for more effective brainstorming but also for other type of ideation.

Idea challenge is a focused form of innovation where you raise a problem or opportunity with the hopes of coming up with creative solutions.

The point of idea challenge is to participate in ideation and generate ideas around a pre-defined theme for a limited period of time.

It allows you to form a specific question and direct that question at a specific audience to receive new ideas and unique insights.

Before setting up an idea challenge, it’s important to define what you want to accomplish with it.

Because there are two types of idea challenges, problem centric and solution  centric approaches, you should first clarify whether you’re looking to identify challenges or develop potential solutions for them.

When organizing an idea challenge, there are different parameters that you can choose to achieve the outcomes you’re looking for, such as theme, audience, responsibilities, time, or channels .

Keep in mind that idea challenge is the best technique when you need to generate lots of new ideas. It may not be the most effective way to generate ideas if you only involve a few experts in your ideation process as it’s proven to be more useful for engaging large audiences.

Although idea challenge enables you to gather lots of ideas fast, careful planning takes time and might not be worth the effort if there are no resources to execute it properly. Also, right timing is necessary for it to succeed.

The SCAMPER technique is created by Bob Eberle, and is a method used for problem-solving and creative thinking. It’s a holistic way of applying critical thinking to modify ideas, concepts or processes that already exist.

The purpose of the SCAMPER is to make adjustments to some parts of the existing idea or process to reach the best solution. It consists of seven actions that can be used to replace parts in the process:

  1. Substitute  – Substitution technique refers to replacing a part of your product, concept or process with another to achieve even better outcome.

2. Combine  – The combine technique explores the possibility to combine two ideas into a single, more effective solution.

3. Adapt  – Adaptation analyses the possibilities to make the process more flexible and focuses on other similar incremental improvements to the idea, process, or concept.

4. Modify  – Modifying the idea looks at the problem or opportunity from a bigger perspective and aims for improving the overall results, not just the idea.

5. Put to another use  – This approach focuses on finding ways to use the idea or existing solution for another purpose and analyses the possible benefits if applied to other parts of the business.

6. Eliminate  – The elimination technique is quite straightforward: it examines the possible outcomes if one or more parts of the concept were eliminated.

7. Reverse – This action focuses on reversing the order of interchangeable elements of an idea.

Although the SCAMPER technique was originally designed for brainstorming sessions, it can be applied to your own thought processes as well.

Often, people tend to focus on finding the next big idea. When generating new ideas, it is easy to forget that the continuous incremental improvements are the ones that really make an impact in the long run.

When you make ideation a constant practice, you’ll have a good chance to win that big idea through a number of small ones. Sometimes, all it may take is to have look at what you’ve already got. Using your existing ideas or processes as a starting point can reveal a lot about your current situation, which is what the SCAMPER technique is about.

Opposite/reverse thinking is a technique that can help you  question long-held assumptions  related to your business. It’s a useful tool to consider if you feel your team is stuck with the conventional mindset and coming up with those “out-of-the-box ideas” seems to be difficult.

Often, finding the best solutions aren’t found through a linear thought process. Although our brains are wired that way, opposite thinking can help us question the norm.

With this type of thinking, you consider the exact opposite of what’s normal. You can even think backwards to find unconventional solutions.

Brainstorm cards are a useful tool created by the Board of Innovation for  coming up with dozens of new ideas  related to whatever challenge or problem you are currently working with. Brainstorm cards help you consider external factors such as: societal trends, new technologies, and regulation in the context of your business.

Brainstorming cards can be used as part of more complex workshops, as brainstorming sessions take different shapes and forms.

We wrote an in-depth guide on how to run an efficient brainstorming session and how to make the most out of it. You can also prepare for brainstorming by collecting ideas in Viima, which you can later discuss during your in-person session.

Analogy thinking  is a technique for using information from one source to solve a problem in another context. Often one solution to a problem or opportunity can be used to solve another problem.

Analogy thinking can, for example, be used for analyzing a successful business, identifying what makes it great, and then applying those same principles for your business. This is an effortless method for coming up with new ideas that are pre-validated .

You’ve probably heard of the countless start-ups that are the “Uber for  [insert industry here] “. This is exactly the method every one of those companies has used. However, although this is such an easy and intuitive tool to use, the obvious combinations are likely to be very competitive. The aforementioned techniques are some of our favorites for generating ideas but definitely not the only ones out there . The ones that work for you can be found by testing different options.

Managing ideas

Although generating ideas is often the easy part of innovation, collecting and managing them can be challenging without a proper tool.

Because people often come up with new ideas then and there, they should be able collect these ideas right when they arise.

Many people use digital or physical notebooks to write down their ideas on the go. This method works well when you just want to make notes for your own use. However, when you want to share that information with others and ask for feedback for further improvement, this isn’t the most effective way.

If you need to involve more people in your ideation process , you might want to consider a dedicated idea management tool to make sure all of the potential ideas are collected in one place.

As idea management is such a vast topic and larger companies are always going to have thousands of ideas, using a dedicated idea management tool  often makes sense for most companies.

An idea management tool serves as the backbone of your idea management process. It's where you’ll collect the ideas, discuss them, evaluate and prioritize them, and keep track of their progress, as well as the strategic direction of your ideation activities.

Managing ideas via a dedicated tool is only as effective as the underlying process behind it. To make ideation a continuous habit , you should build a process that makes generating and refining new ideas easy. Too complex processes will only frustrate users, so don't try to make it too complicated.

If you want to get started with an idea management tool, book a demo with Viima and see the software in action.

Succeeding with Idea Generation

Before engaging more people to use these tools and techniques for generating more ideas, there are a few more tips that can help you get more out of your ideation methods.

Although this seems obvious, the more accurately you’re able to describe your current or perceived problem or opportunity, the better chances you have of actually generating useful ideas.

Before you start, try to gather as much information about the problem as possible to get to the heart of the problem. Identify what you already know about it and what information is still needed.

For your ideation to be as productive as possible, it is important to involve the right people to the process. Engage those people who know about the topic and are genuinely interested in making a change in your organization.

If your plan is to engage a larger group of people to generate ideas, make sure your audience is relevant and well informed on the subject.

Communicating that any idea is a good idea may hinder creativity, so make sure your goals are ambitious and specific enough. One way to get more of actually creative ideas from your audience is to set constraints.

If your general goal was to reduce costs, it’s likely that people would suggest ideas such as spending less on office equipment or buying cheaper coffee. However, the ideas you’ll get will be a lot different if you’d ask: “What could we do to save 70% on costs?”.

Without constraints, people typically come up with small suggestions for improvement with little or no creativity. But when describing the challenge in more detail and setting more ambitious goals, people really need to use their imagination to achieve a desired outcome.

The purpose of generating new ideas is about improving what already exists as well as coming up with something new.

Often when trying to solve a problem or approaching a new opportunity, people use existing solutions or mental models instead of trying to come up with new ones. The problem with this approach is that it doesn’t allow you to explore different opportunities and it narrows down the number of possible solutions.

Coming up with completely new ideas can help you approach your problem or opportunity from a new perspective. It enables you to expand the range of ideas beyond your current way of thinking which eventually leads to more ideas.

There are several different methods and techniques that can be used to challenge that conventional mindset and to get more of those high-quality ideas. The ones introduced in this post can be used as a reference in your brainstorming sessions or when generating ideas on your own.

The next step towards generating more ideas would be to try some of these methods in practice. You can download our  idea generation toolkit  that consists of some of our favorite tools for idea generation.

This post is a part of our  Idea Management   blog-series. In this series, we dive deep into the different areas of idea management and cover the aspects we think are the most important to understand about managing ideas.

You can read the rest of the articles in our series covering idea management by clicking on the button below. Don’t forget to subscribe to our blog to receive updates of our upcoming content!

Related articles

The Complete Guide to Ideation

Idea Management – The Complete Guide

Innovative Thinking: Why It's The Skill of the Future

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Clin Epidemiol

From ideas to studies: how to get ideas and sharpen them into research questions

Jan p vandenbroucke.

1 Leiden University Medical Center, Leiden, the Netherlands

2 Department of Clinical Epidemiology, Aarhus University, Aarhus, Denmark

3 Department of Medical Statistics and Centre for Global NCDs, London School of Hygiene and Tropical Medicine, London, UK

Neil Pearce

Where do new research questions come from? This is at best only partially taught in courses or textbooks about clinical or epidemiological research. Methods are taught under the assumption that a researcher already knows the research question and knows which methods will fit that question. Similarly, the real complexity of the thought processes that lead to a scientific undertaking is almost never described in published papers. In this paper, we first discuss how to get an idea that is worth researching. We describe sources of new ideas and how to foster a creative attitude by “cultivating your thoughts”. Only a few of these ideas will make it into a study. Next, we describe how to sharpen and focus a research question so that a study becomes feasible and a valid test of the underlying idea. To do this, the idea needs to be “pruned”. Pruning a research question means cutting away anything that is unnecessary, so that only the essence remains. This includes determining both the latent and the stated objectives, specific pruning questions, and the use of specific schemes to structure reasoning. After this, the following steps include preparation of a brief protocol, conduct of a pilot study, and writing a draft of the paper including draft tables. Then you are ready to carry out your research.

Introduction

How do you get an idea for a study? How do you turn your idea into a testable hypothesis, and turn this into an appropriate and feasible study design? This is usually at best only partially taught in epidemiology courses. Most courses and textbooks assume that you know your research question and the general methods that you will need to answer it. Somehow it is assumed that you can readily translate your idea into a specific framework, such as the PICO framework (Patient, Intervention, Control or Comparison, Outcome) 1 or the FINER framework (Feasible, Interesting, Novel, Ethical, and Relevant) 2 or that you can fit it into counterfactual reasoning. 3 However, before describing your project in one of these frameworks, you first need to have an idea for your study and think about it in general terms: why you might do a study and how you might do a study.

This paper considers the complex process of having ideas, keeping track of them, turning them into studies, trying them out in pilot studies, and writing a draft paper before you finally embark on your study.

The paper is intended for novice researchers in clinical or public health epidemiology. It is not intended to be a comprehensive literature review about creativity, nor a sociology or philosophical treatise about why scientists get particular ideas (and not other ideas). It is based on our personal experience of (a combined) 70+ epidemiologic research-years. We have worked on very different topics, mostly on opposite sides of the globe, yet found that our experiences are quite similar. The fact that these issues are rarely covered in epidemiology courses has provided motivation to reflect on our experience.

Getting new ideas

So how do you get an idea? How some juxtaposition of neural patterns in our brain suddenly creates a new idea is a process that we are far from understanding. According to Karl Popper, the origin of new ideas does not matter; the only thing of interest is to devise how to test them. 4 Over the past decades, the literature has been enriched with new ideas about “being creative” in science – as witnessed in the book Innovation Generation by Ness. 5

In the present paper, we will not cover the literature about creativity and discovery in depth, but we will discuss the issues that we consider relevant to epidemiologic research. We will first consider the more general principles.

The real complexity of the thought processes that lead to a scientific undertaking is almost never described in published papers. Immunologist Medawar claimed that in this respect almost all scientific papers may be a fraud – not in the sense that scientists deliberately produce misleading data, but in the sense that the real thought processes that lead to the data and conclusions are not mentioned. 6 Scientists tell us about their real thought processes in memoirs, inaugural, or valedictory lectures – which is why these are so much more interesting than “standard” papers or presentations.

What strikes our minds: regularities or anomalies?

All sciences study a particular “object of knowledge” (eg, “matter”, “life”). Ideas come from experience and previous knowledge or facts about this object of knowledge, although this knowledge is always filtered through the perspective of one or more theories. 7 Epidemiology studies the distribution and determinants of disease in human populations, 8 and epidemiological ideas arise from observing and thinking about populations. 9 These could be clinical populations (ie, clinical experience, sometimes involving just a few patients), exposure-based populations (eg, workers exposed to a particular chemical), or general populations (geographically defined or sociologically defined). Whatever the population we are interested in, ideas come from observing either regularities or anomalies.

The observation of regularities (“induction”) is a common origin of new ideas. 4 , 10 – 13 Philosopher David Hume described “Induction” as: regularly seeing two things happening in succession (like pushing a switch and a light going on) leads to suspicions of causality. As he pointed out, causality can never be proven by the mere observation of “constant conjunctions”, but observing regularities can start our train of thought. 12

An anomaly (or irregularity) strikes our mind, because it defies our expectations. The regularity that we expected was our “hypothesis” (even if it was not really explicitly formulated); the anomaly is a “refutation”. 4 , 13 It forces us to think about other explanations, and these lead to new hypotheses that we then try to test. Thus, scientists do not usually start from hypotheses that are nicely formulated “out of the blue”, but instead start from previous knowledge and experience; when they are challenged by anomalies, scientists seek new explanations. 14

An interesting way to discover anomalies is to enter a new field of research; since you have other background experience than the people already in the field, you see things that they take for granted but that strike you as odd – at the same time, you may also see new explanations for these anomalies. One of the pioneers of clinical epidemiology, Sackett, once wrote that scientists should “retire” from a field as soon as they become “experts”. 15 When you are too long in a field, you will no longer see the anomalies, and you may even obstruct newcomers with new explanations. Of course, there are differences between scientists: some roam across various fields and others stick to a problem area that they explore with increasing depth – then the increasing depth and the new techniques that one needs for advancing one’s thoughts will be like a “new field”.

Taxonomies of discovery

Few researchers have listed the different ways in which one can arrive at new ideas, that is, lists of ways of discovery. We will present two of them – which have very different origins but remarkable similarities. Several examples of studies corresponding to items on these two lists are given in Appendix Examples A1–A10 .

Sources for new ideas about health care evaluation were described by Crombie and Davies in the chapter “Developing the research question” of their book on Research in Health Care that reflects a UK public health experience. 16

• “Review existing practice […] the current organisation and delivery of health care is not as good as it could be […]”
• “Challenge accepted ideas […] much of health care is based on accepted practice rather than research evidence […]” ( Appendix Example A3 )
• “Look for conflicting views […] which indicate either that there is not enough evidence, or that some practitioners are misinformed”
• “Investigate geographical variation […] reflecting on the reasons [for geographical variation] can be a fruitful source of research questions […]” ( Appendix Example A6 )
• “Identify Cinderella topics […] important areas of health care are often overlooked […]”
• “Let loose the imagination […] look for wild or impossible ideas […] free the mind from the constraints of conventional wisdom […].”

A taxonomy for sources of clinical research questions about medical care and clinical problems was proposed by Hulley and Cummings, in the context of clinical research in the US: 2

• “Build on experience;” your own experience, that of close colleagues with whom you can freely discuss your research ideas, and that of a good mentor, because young researchers might not yet have much experience, “An essential strategy for a young investigator is to apprentice himself to an experienced senior scientist who has the time and interest to work with him regularly.”
• ○ By harvesting “the medical literature and attending journal clubs, national and international meetings, seeking informal conversations with other scientists and colleagues”
• ○ “A sceptical attitude about prevailing beliefs can stimulate good research questions”
• ○ Be alert to “careful observation of patients, which has historically been one of the major sources of descriptive studies” ( Appendix Examples A1 and A2 )
• ○ Your experiences in teaching; having to explain something may make you aware of gaps in your knowledge; questions by patients and colleagues may similarly identify things that we do not fully understand or ignore
• “Keep the imagination roaming […]” by a mixture of creativity and tenacity; “put an unresolved question clearly in view and turn on the mental switch that lets the mind run freely toward it”.

A special mention needs to be made about the last categories of both the lists: “Let loose the imagination” and “Keep the imagination roaming”. These are especially important to find innovative solutions. In many situations wherein you cannot do a perfect study and you run a grave danger of potential confounding or bias, it helps to “get deeply immersed”: to understand the problem biologically, clinically, socially, organizationally, and environmentally will help you to think about what is happening, why it is happening, and whether you can find situations in which the potential confounders or biases do not exist or exists in reverse. You should forget formal designs and think out of the box: you will find instances of studies that mutually reinforce each other and may even arrive at formulating new designs or analytic solutions (see Appendix Examples A7–A10 ).

Keeping track of your ideas

It is not only important to have good ideas but also important to develop them. Researchers who work in laboratories have the habit of keeping “lab logs”. They write down briefly the results of an experiment, note why they think it went wrong, and how they will perform the next experiment. This permits them to trace how they changed the experiments or even the content and the direction of their research. We should do the same in epidemiologic and clinical research, particularly in the stage of creating new ideas. Such notes about ideas can include not only hypotheses and views or results by others but also drawing directed acyclic graphs (DAGs) (see “Intermezzo: specific schemes to structure reasoning” section) to make the causal structures of ideas clear.

The greatest minds kept track of their thoughts. Charles Darwin’s notebooks document his ideas, his observations, his readings, and new theories and facts that struck him. 17 For example, Darwin noted a story that he heard from his father, a medical practitioner. His father recounted that he had been struck by one of his patients’ ways of expressing himself, because he had attended a parent of the patient who had had the same mannerisms – even though the parent had died when the patient was still an infant. Remarks like these still have relevance today when we think about the heredity and evolution of behavior.

The sociologist C Wright Mills carried the description of the process one step further in the appendix of his book on The Sociological Imagination . 18 He encourages young sociologists to set up a file of stacked cards to keep track of “[…] personal experience and professional activities, studies underway and studies planned […]” which “[…] encourages you to capture ‘fringe thoughts’: various ideas which may be by-products of everyday life, stretches of conversations […]”. These notes are continuously reshuffled, regrouped under new headings, and pondered. Mills denounced the habit of most (social) scientists who feel the need to write about their plans only when they are going to apply for a grant. He thought that scientists should continually work with their file of ideas and regularly take stock of how these have evolved.

Such strategies are still relevant today, even if our “logs” are kept in electronic form, particularly because grant writing has become more demanding, hectic, and time-consuming. From such files, new research projects are born: while your ideas gradually develop, you keep wondering what data you might need to prove a certain proposition, and how you might get those data in the easiest way possible. Often, ideas are reshuffled and regrouped under new headings. A new observation, a new piece of literature may make old ones fall into place, or there may suddenly be a new opportunity to work out an old idea.

A complementary advice recently came in a blog from a contemporary sociologist, Aldrich: his advice is to “Write as if you don’t have the data”, that is, to write “[…] the literature review and planning phase of a project, preferably before it has been locked into a specific research design”. 19

The role of emotions

Underlying the discovery process, there are often two emotions: “surprise” and “indignation”. Surprise is the intellectual emotion when we see something happening against expectation: a patient with an unusual exposure, unusual disease manifestation, sudden cure, or sudden ill-understood deterioration; a laboratory result that is an anomaly; and a sudden epidemic of disease in a population. Indignation is the moral emotion: a group of patients is not being treated well because we lack sufficient knowledge, or because we are blundering in organizing health care or in transmitting and applying public health knowledge. Some passion is useful to bring any undertaking to a good end, be it that the passion should be restrained and channeled into polite undertakings, like in a research protocol. While doing the research project, maintaining some of the original passion will help you to find ways to overcome the daily hassles of research, the misadventures, the difficulties of getting others to collaborate, and the difficulties of getting published ( Appendix Example A11 ).

Sharpening the research question: the pruning

Pruning a research question means cutting away anything that is unnecessary, so that only the essence remains.

The initial spark of an idea will usually lead to some rather general research question. Invariably, this is too ambitious, or so all-encompassing that it cannot be researched (at least not within the time frame of a single grant or PhD project). You have to refine your research question into something that is interesting, yet feasible. To do so, you have to know clearly where you are heading. The emphasis on a clear preconceived idea about what you want to attain by your research often comes as a surprise; some people object: “[…] isn’t research about discovery? How can you know in advance what you want to find?”

The social scientist Verschuren proposed the “wristwatch metaphor”. 20 A researcher is not like a beachcomber, who strolls along the beach to see whether anything valuable washed ashore. Rather, a researcher is like someone who has lost her wristwatch on the beach and returns to search for it. She knows what part of the beach to look, she can describe her wristwatch in detail, and once she has found it, she knows that this is the watch she was looking for. Some further background to these ideas can be found in Appendix B .

Charles Medawar wrote in his Advice to a Young Scientist (page 18) 21 that as much as politics is the ‘art of the possible’, research is the ‘art of the soluble’. A research question should be limited to a question that can be solved with the resources at hand. This does not mean that you should preferentially study “trivial” questions with easy solutions. It does mean that you should seek out your particular niche: something specific, something that was overlooked by others, or some new twist to a general question, so that you can make your own contribution.

The concept of “serendipity” is often invoked when thinking of “seeking novelty”: it means finding something that you were not looking for. For a full discussion of the more complex reality that shows how, in reality, “chance favors a prepared mind”, see Appendix C .

Proceed in the inverse order of the paper that you will write

From the aforementioned, we know that we need a precise aim and a soluble research question.

How can we achieve this? The best approach is to “begin at the end”, that is, the conclusion that you hope to support when you eventually publish your research findings, perhaps many years from now. 22 Most medical research papers have a fixed format: introduction, methods, results, discussion. Usually, the discussion has three parts: summary of the results, discussion of the strengths and limitations, and the importance and interpretation of the findings. There you start: you try to imagine what such last lines of the eventual paper might be – in particular what their intent, their message to the reader might be. Another useful strategy would be to imagine what might be written in the separate box “What this paper adds” that many journals nowadays ask to convey the message from the authors clearly and succinctly to the readers.

The “latent” versus the “stated” objective

The pioneer clinical epidemiologist Feinstein wrote that a good research consultant should be like a good clinician, who first wants to learn from the patient: “What is the chief complaint?”, that is, which is the problem that you want to study. Next, “What will you do with the answer?” 22 The latter question is not just about the potential conclusions of the research paper, but more importantly, their meaning. What is the intended effect (or impact) of the findings? He called this the “latent objective”: what do you want to achieve or change by your project; the “stated objective” is different, it is the type of result that the study will deliver. For example, the stated objective can be that you want to do a randomized trial to compare one intervention versus another and that you will look at recurrence of disease. The latent objective might be that you are concerned that one intervention may be harmful to patients, driven by special interests, and that if this is the case it should be abolished.

Rather analogously, the long-time editor of the Annals of Internal Medicine , Edward Huth, proposed in his book about medical publishing the “So-What” and the “Who-Cares” tests: “What may happen if the paper’s message is correct?”; may it change concepts and treatment or stimulate further exciting research? 23 In fact, many funders now require such an “impact statement” as part of the grant application process.

Experienced research consultants know that when trying to discover the latent objective, it is useful to brush aside the detailed protocol and to ask directly what the meaning of the research is. The meaning of the research is often not clearly stated in a formal study protocol that limits itself more or less to “stated aims”. 24 Like a patient who cannot articulate her/his complaints very well, would-be researchers lose themselves in trivial “side issues” or operational details of the protocol. Appendix Examples A2 and A11 explain the importance of elucidating the underlying frustration of the clinician-researcher to clearly guide a research effort.

After initial questions have set the scene and clarified the “latent objective” of a project, the next questions are more operational, translating the latent objective back into a “stated objective”. 22 The stated objective should be a feasible research project. According to Feinstein, one should ask: what maneuver is to be executed (what intervention, deliberate or not, and how is it administered), what groups are to be compared (and why those groups), and what is the outcome that we will study?

In these phases of discussion, one needs to immerse oneself into the problem: one has to understand it biologically and clinically, and how it is dealt with in the daily practice of health care in the setting in which you will do research. Getting deeply immersed in the problem is the only way of arriving at shrewd or new solutions for studies on vexing medical or public health problems ( Appendix Example A9 ). Mere discussion of technical or procedural aspects of a proposed design, data collection, or analysis will usually not lead to new insights.

Specific pruning questions, to ask yourself or others

In initial discussions, one goes back and forth between the general aim (the latent objective), the scientific questions that follow from it, and the possible research designs (with stated objectives). After feeling secure about the “latent” aim, proceed with more specific questions.

• Try to describe exactly the knowledge gap that you want to fill (ie, the watch that you lost at the beach). Is it about etiology, about pathogenesis, about prognosis? What should change for the benefit of a particular group of patients? Try to be as specific as possible. Do your colleagues see these problems and their solutions as you do? – and if not, why don’t they?
• Once you know the point you want to make, describe what table or figure you need to fill the gap in knowledge, that is, what would your results look like? This means drawing a simple table or graph. Are these the data you want? Will these tables convince your colleagues? What objections might they have? Keep in mind that if the research results go against ingrained beliefs, they will be scrutinized mercilessly, so the important aspects of your research should be able to withstand likely objections.
• Thereafter, the questions become more practical: what study design is needed to produce this table, this figure? Can we do this? Do we have the resources or can we find them?

Be self-critical

You should always remain self-critical about the aspects that threaten the validity of your study ( Appendix Example A12 ). 25 If the practical problems are too large, or the research question too unfeasibly grandiose, it might be wise to settle for a less ambitious aim ( Appendix Example A13 ).

Paraphrasing Miettinen, 26 the first decision is whether you should do the study at all. There might be several reasons to decide not to pursue a study. One might be that arriving at a satisfactory design will be impossible, because of biases that you are unable to solve. It serves no purpose to add another study that suffers from the same unsolved problems as previous studies. For example, it does not serve any purpose to do yet another study that shows lower mortality in vegetarians, if you cannot solve the problems of confounding that vegetarians are persons who have different lifestyles in comparison with others. 27 (If, however, you have found a solution – pursue it at all means!) Nevertheless, thinking about the potential problems and ultimate aims of a seemingly impossible question can foster the development of a new study design or a new method of analysis, ( Appendix Examples A2, A9, and A10 ). In the same vein, deciding that you cannot do a study yourself might make you look for collaboration with persons who have the type of data that you do not, for example, in a different population where it is believed that confounding is not so severe or may even be in the opposite direction.

All studies have imperfections, but you need to be aware which ones you can tolerate. 28 In the early stages of an enquiry, an “imperfect” study might still be worthwhile to see whether “there might be something in it”. For example, time trends or ecological comparisons are often seen as poor study designs to assess causality by themselves, but they can be very valuable in helping to develop ideas, as well as providing a “reality check” about the potential credibility of some hypothesis. 29

Conversely, it is pointless to add yet another study, however perfect, showing what is already known very well – unless you have to do it for “political” purposes, say, for convincing decision makers in your own country.

Finally, it is not a good use of your time to chase something completely improbable or futile. For example, at the present state of the debate, it serves no purpose to add another study about the presence or absence of clinical benefits or harms of homeopathy: no one will change his or her mind about the issue. 30 , 31 An exception might be something that is highly improbable, but that if true might lead to completely revolutionary insights – such an idea might be worth pursuing, even if the initial reaction of outsiders might remain incredulousness. Still, you should pursue unlikely hypotheses knowingly, that is, with the right amount of self-criticism – in particular, to make yourself aware when you are in a blind alley.

To keep yourself on the “straight and narrow”, it helps to form a group of people who cover different aspects of the problem you want to study: clinical, biochemical and physiological, and methodological – to discuss the project as equals. Such discussions can not only be tremendous fun but also will invariably lead to more profound and diverse research questions and will help to find solutions for practical as well as theoretical problems. In the right circumstances of a “machtsfreie Dialog” 32 (a communication in which all are equal and that is only based on rational arguments and not on power – which all scientific debates should be), such a circle of colleagues and friends will help you to be self-critical.

Finally, when pursuing one’s research interests, one should be prepared to learn new skills from other fields or collaborate with others from these fields. If one stays only with the techniques and skills that one knows, it might not lead to the desired answers. 33

What if the data already exist? And you are employed to do a particular analysis with an existing protocol?

Even in the circumstance that the data already exist, it greatly helps to not jump into an analysis, but to think for yourself what you would ideally like to do – if there were no constraints. As Aldrich mentioned, 19 also in that circumstance researchers should still

[…] begin their literature review and conceptual modeling as if they had the luxury of a blank slate […]. Writing without data constraints will, I believe, free their imaginations to range widely over the realm of possibilities, before they are brought to earth by practical necessities.

Moreover, this will make clear what compromises one will make by accepting the available data and the existing analysis protocol. Otherwise, one starts an analysis without being sufficiently aware of the limitations of a particular analysis on particular data.

The difference between explanatory and pragmatic research

A useful distinction is between explanatory and pragmatic research: the former is research that aims at discovery and explanation, whereas the latter is intended to evaluate interventions or diagnostic procedures. The first type of research consists of chasing explanations by pursuing different and evolving hypotheses; the second type of research aims at making decisions about actions in future patients. 27 The two opposites differ strongly in their thinking about the types of studies to pursue (eg, observational vs randomized), about the role of prior specification of a research hypothesis, about the need for “sticking to a prespecified protocol”, and about subgroup analyses and multiplicity of analyses. Some of these will be explained in the following subheadings.

The difference between explanatory and pragmatic trials is sometimes thought to mirror the difference between doing randomized trials versus observational research. However, even for randomized trials, a difference exists between “ pragmatic” and “explanatory” trials (coined first by Schwartz and Lellouch). 34 Because it is not always easy to delineate what aspects of a randomized trial are “pragmatic” or “explanatory”, instruments have been crafted to help researchers and evaluators. 35 , 36 Conversely, not all observational studies are explanatory: some are needed for pragmatic decisions (think about adverse effects of drugs and also about diagnostic evaluations where studies should influence practice guidelines) – while other studies aim at explaining how nature works.

Which iterations should you allow yourself? Anticipating the next project

Thinking about a research problem is a strongly iterative process. 2 , 33 , 37 One starts with a broad aim and then tries out several possible ideas about studies that might lead to better understanding or to better solutions.

Likewise, project proposals characteristically go through many iterations. In the early phases of the research, it is commonplace that the study design or even the research question is changed. Specific suggestions about common research problems and their potential solutions were given by Hulley and Cummings, 2 which we reproduce in Appendix D .

The revision of the aims of a project may be profound, in particular in explanatory research (see “The difference between explanatory and pragmatic research” section), in contrast to pragmatic research (see “Shouldn’t you stick to a predefined protocol?” section). The chemist Whitesides wrote: “Often the objectives of a paper when it is finished are different from those used to justify starting the work. Much of good science is opportunistic and revisionist”. 38 Along a similar line, Medawar proposed that to do justice to the real thought processes of a research undertaking, the discussion section of a paper should come at the beginning, since the thought processes of a scientist start with an expectation about particular results. The expectation determines which findings are of interest and why they will be interpreted in a particular way. 6 He added that in real scientific life, scientists get new ideas (ie, new expectations) while doing their research, but “[…] many of them apparently are ashamed to admit, that hypotheses appear in their mind along uncharted byways of thought”. 6

“Seeing something in the data” can be an important part of scientific discovery. This is often decried as “data dredging”, which it is not: one sees something because of one’s background knowledge and thereby there always is some “prior” that exists – even if that was not specified beforehand in the study protocol. 27 , 39 The word “exploratory” is often misused when it is used to characterize a study. True “exploratory” data analysis would only exists if it is mindlessly done, such as a Genome Wide Association Study (GWAS) analysis – but even GWAS analyses have specific aims, which becomes clear when results are interpreted and some findings are designated as “important” and others not. As stated by Rothman:

Hypotheses are not generated by data; they are proposed by scientists. The process by which scientists use their imagination to create hypotheses has no formal methodology […]. Any study, whether considered exploratory or not, can serve to refute a hypothesis. 40

Appendix Examples A5 and A7 show how projects changed mid-course because of a new discovery in the data or in the background knowledge about a research topic.

Generally, it is a good habit to think through what the next project might be, once you will have the result of the project you are currently thinking about, so as to know what direction your research might take. 33

Shouldn’t you stick to a predefined protocol?

Different research aims, in particular along the “explanatory” versus “pragmatic” continuum, may lead to different attitudes on the amount of change that protocols may endure while doing research. 27 , 39 For randomized trials, and also for pragmatic observational research, the research question is usually fixed: does a new therapy lead to better outcomes for a particular group of patients in a particular setting? Because findings from randomized trials or pragmatic observational research may lead to millions of patients to adopt or avoid a particular therapy (which means that their well-being or even life depends on the research) researchers are generally not at liberty to change their hypotheses at the last moment – for example, by suddenly declaring an interest in a particular subgroup. They should stick to the predefined protocol. If a change is needed for practical reasons, it should be clearly stated in the resulting publications. This makes thinking about research questions and doing pilot studies beforehand all the more important (see “Pilot Study” section).

In contrast, much epidemiologic and clinical research tries to explain how nature works. This gives greater leeway: exploration of data can lead to new insights. Thus, “sticking to the protocol” is a good rule for randomized trials and pragmatic observational research, but may be counterproductive for explanatory research. 39 , 41 Nevertheless, it is good to keep track of the changes in your thoughts and in the protocol, even if only for yourself. In practice, many situations are intermediate; in particular when using large available data sets, it often happens that one envisages in a protocol what one would do with the data, only to discover upon opening the data files that the data fall short or are more complex than imagined; this is another reason for doing pilot studies, even with large available data sets (see “Pilot Study” section).

How much literature should you read?

If you are setting up a new research project in a new area, do not start by reading too much. You will quickly drown in the ideas of others. Rather, read a few general reviews that identify unanswered problems. Only return to the literature after you have defined your research question and provisionally your study design. Now, the literature suddenly becomes extremely interesting, since you know what types of papers you need. You also know what the potential objections and shortcomings are of the different design options, because you thought about them yourself. The number of relevant papers usually greatly shrinks, see Appendix Example A4 .

Shouldn’t you do a systematic review first?

It is argued that before embarking on a new piece of research, one should first do a systematic review and/or meta-analysis, because this may help to define the gaps in knowledge more precisely, and guide new research – or may show that the question has been solved. This argument is somewhat circular. A systematic review is a piece of research in itself, intended for publication, and requires much time and effort. Like any piece of research, it requires a clear research question. As such it does not “identify gaps”: a systematic review is about a research question which is already specified, but for which more information is needed. Thus, the main function of the advice to first do a systematic review is to know whether the research question that one has in mind has not yet been solved by others. Perusing the literature in depth is absolutely needed, for example, before embarking on a randomized trial or on a major observational study. However, this is not the same as doing a formal systematic review. In-depth scoping of the literature will suffice. If it is found that potentially valuable studies already exist on the research question that one has in mind, then the new study that one is thinking about may be discarded, and a systematic review should be done instead.

Intermezzo: specific schemes to structure reasoning

Specific schemes have been proposed to guide our reasoning between the stage of delineation of the “gap in knowledge” and the stage of proposing the research design.

The acronym FINER (feasible, interesting, novel, ethical, and relevant) was coined by Hulley and Cummings 2 and denotes the different aspects that one should consider to judge a budding research proposal. These words are a good checklist for an in-depth self-scrutiny of your research. The central aspects are the feasibility and whether the possible answers are exciting (and/or much needed).

The PICO format (Patient, Intervention, Control or Comparison, Outcome) is advocated by the evidence-based medicine and Cochrane movements and is very useful for clinical therapeutic research, particularly randomized controlled trials (RCTs). 1 , 42 Questions about therapeutic interventions are highly specific, for example, a particular chemotherapeutic scheme (the intervention) is proposed to study survival (the outcome) among young women with a particular form of stage III breast cancer (the patients). This framework is less useful, and becomes a bit pointless, for etiologic research about generalizable questions such as: “Does smoking cause lung cancer?” which applies to all humans and to different types of smoking. Of course, all research will be done in particular population, with particular smoking habits, but this does not necessarily define the research question. Some of the first investigations about smoking and lung cancer were done in male doctors aged ≥35 years in the UK 43 – this was a very convenient group to research, but being a male doctor in the UK is not part of the research question.

The PICO format is thus most applicable for pragmatic research. A much more detailed and elaborate scheme for pragmatic research was proposed by the US Patient-Centered Outcomes Research Institute (PCORI) which has published Methodology Standards, including “Standards for Formulating Research Questions”. While we would not agree with all six standards, junior investigators may find the structure useful as they think through their options – especially for pragmatic research questions. 44

Counterfactual reasoning 3 emphasizes those aspects of the “ideal randomized trial” that should be mimicked by an observational study. A key question is whether your study is addressing a hypothesis that could in theory be studied in a randomized trial. For example, if the research question is “does smoking cause lung cancer?”, then this is a question that could in theory (but not in practice) be addressed by randomizing study participants to be smokers or nonsmokers. In this situation, it may be useful to design your observational study with the intention of obtaining the same answer that would have been obtained if you had been able to do a randomized trial.

However, the aims of explanatory observational research are different from those of randomized trials. 27 Explanatory research about disease etiology may involve “states” like being female, being old, being obese, having hypertension, having a high serum cholesterol, carrying the BrCa1 gene, and so on, as causes of disease. None of these causes are interventions. In contrast, RCTs focus on what to do to change particular causes: which interventions are feasible and work? For example, being female might expose a person to job discrimination; the intervention might be to have women on the appointment committee or to use some kind of positive discrimination. Likewise, the gene for phenylketonuria leads to disease, but the intervention is to change the diet. For carriers of BRCa1 genes, different strategies can be evaluated in RCTs to evaluate their effectiveness in preventing premature death due to breast cancer: frequent screening, prophylactic mastectomy, hormone treatment, and so on – which may have different effects. For obesity or hypertension or hypercholesterolemia, different types of interventions are possible – with potentially different effects and different adverse effects.

The interventionist outlook, that is, trying to mimic an RCT, can be very useful, for some type of observational studies, for example, about the adverse effects of drugs. It helps to make certain that one can mimic an “intervention” (ie, patients starting to use particular drugs) that is specific and consistent in groups of patients that are comparable (more technically, exchangeable – meaning that the results of the investigation would not change if the persons exposed and nonexposed were swapped). These conditions can be met in a credible way, if there are competing drugs for a similar indication, so that there is an active drug comparator: the interventions (use of different drugs in different patients) will be well defined, and the patients on the different drugs will tend to be comparable. This works particularly well if you are focusing on adverse drug effects that were unknown or unpredictable at the time of prescription. 45 , 46 For example, you may obtain more valid findings in a study that compares the adverse effects of two different beta agonists for asthma care (ie, two different drugs within the same class), than to design a study which compares patients who are prescribed beta agonists with patients who are prescribed other asthma medication, or no medication at all – because the latter might be a highly different group of patients. 47

As mentioned, there are some important studies about causes of diseases where a randomized trial is not feasible, even in theory. In particular, there are various “states” which are major causes of disease (obesity, cholesterol, hypertension, diabetes, etc). These states strongly affect the risks of disease and death, but cannot be randomized. For example, it is difficult to conceive of randomizing study participants to be obese or not obese; however, we could randomize them for the reduction of obesity, for example, through exercise, but such a study would assess the effects of a particular intervention, not of obesity itself. Still, it remains important to estimate the overall effects of obesity, that is, to answer the question “would this group of people have had different health status, on the average, if they had not been obese”. In this situation, the concept of “interventions” is not relevant to designing your study (at least in the way that the term “intervention” is commonly used). What is more relevant is simply to focus on the counterfactual contrast which is being assessed (eg, a body mass index [BMI] of 35 versus a BMI of 25), without specifying how this contrast came about.

A technique that has gone hand in hand with counterfactual reasoning in epidemiology is drawing DAGs; several introductions to DAG theory can be found in epidemiologic textbooks. 3 , 48 DAGs can be useful in the brainstorming phase of a study, after the general research question has been defined. At this stage, a general structure for the study is envisaged and the complexity of the causal processes needs clarification. A DAG can be extremely useful for illustrating the context in which a causal question is being asked, the assumptions that will be involved in the analyses (eg, whether a particular risk factor is a confounder, a mediator, or a col-lider), and help us question the validity of our reasoning. 49 Using DAGs helps us also decide which variables we need to collect information on and how they should be measured and defined. Given that DAGs root in causal thinking, their construction is, of necessity, subjective.

Preparation: pilot study, protocol, and advance writing

Doing a pilot study and collecting ancillary information about feasibility.

May I now start? is a question heard after lengthy deliberations about the research question and the potential studies that follow from it. Such deliberations almost invariably produce a lot of enthusiasm and exhilaration – because they are fun. The researcher wants to begin collecting data or start the analysis. However, Crombie and Davies, in their chapter about “Developing the research question” state emphatically: “Don’t rush into a study”. 16 Separate from doing a pilot study, which is about the procedures of your study, you may also need to collect ancillary information before actually starting your study.

Pilot study

Even if you think you are totally certain of what you want, you should first do a pilot study, based on a brief protocol. 2 , 22 That initial protocol should be easy to write. You have already discussed the aim and design of your study. Write them down. You expect a particular type of information that is essential and that will tell the essence of your message (a particular 2-by-2 or X-by-Y table, a particular graph), which you can describe.

Pilot studies are not done to know the likely direction of the results; instead, the aim is to see whether you will be able to perform the procedures of your study – and ultimately whether that really is the study you want to do. 50 The aim is to save yourself from embarrassment: data that very surprisingly do not turn out to be what you expected, questionnaires that are misunderstood or do not deliver the answers that you need or that are not returned, laboratories that do not produce, patients who do not show up, heads of other departments who block access to their patients or materials, or yourself who needs more time to manage the complexity of the undertaking.

We have never heard of someone who was sorry for having done a pilot. Conversely, we know many persons who found out at much personal embarrassment and institutional cost that their project was unfeasible. In intermediate cases, the pilot may show the need to change questionnaires or procedures before the study goes ahead.

In principle, a pilot study should be exactly like your final study and test out all your procedures on a small number of persons. Often, it is better to approach the task piecemeal and pilot different aspects of the research one by one.

A tough question is how to do pilot studies and pilot analyses when ethical or institutional review board approval is necessary for some of the actions in a pilot study. One solution might be to avoid piloting some procedures; for example, try parts of the procedure – for example, you may not be able to randomize in a pilot, but you may be able to try out data collection procedures and forms. There is a degree of circularity about piloting, also in obtaining funding, as one may need funding for the pilot. In practice, the best step might be to ask the ethics committee or review board of your institute which aspects of the research can be piloted and under what conditions.

In Appendix E , several questions that you might ask in pilot studies are listed. They may lead to profound reassessments of your research – particularly if you are piloting the collection of new data, but also if the research involves analyses of existing data.

Ancillary information

It may be necessary to collect additional information about event rates or standard deviations of measurements to calculate the statistical precision that might be obtained. Also, sometimes you need other ways of “testing the water” like procedures to streamlining data collection from different centers in order to know whether the study is feasible. Depending on the study size and importance, such activities may become studies in themselves and actually take a lot of time and money.

Advance writing of paper: before full data collection and/or analysis

Whitesides’ advice is:

The key to efficient use of your and my time is that we start exchanging outlines and proposals as early in a project as possible. Do not, under any circumstances, wait until the collection of data is ‘complete’ before starting to write an outline. 38

After the pilot study, you have a firm grasp of all elements that are necessary for a scientific paper: introduction, materials and methods, results, and discussion. In the introduction, you explain why you have done this research. Almost always, an introduction comprises three ideas: what is the general problem? what is the particular research question? what study will you perform to answer that question? This is followed by the materials and methods section. They have been extensively discussed and have been fine-tuned in the study protocol and the pilot study. Thereafter come the results sections. By now, you know what tables or figures you want and how you can obtain them, but not what the final numbers will look like. You will also have an idea about the auxiliary tables that you might need to explain your data to others (such as a table with the baseline characteristics or an additional table with a subgroup analysis). You can now draft the layouts of all these tables. Visualizing the presentation of your results in advance is the “bare minimum” of writing in advance.

Finally, the discussion section. Can you write a discussion before you know the final data? Of course you can; you even must think ahead. In principle, there are only three possible outcomes: the study can give the results that you hoped for; it can show the inverse; or something indeterminate in between. In all instances, you can imagine how you will react. One possibility is that you are disappointed by the results of your study, and you will tend to find excuses for why it did not produce the results you hoped for. What excuses might your produce? The other possibility is that it does show what you wanted; then you may have to imagine how others will react and what their objections might be. If the results are indeterminate, everybody might be disappointed, and you will need to explain the failure of your research to give clear-cut results. When you detect a specific weakness by imagining this situation, you may wish to change aspects of your study.

As we explain in Appendix F , there is no need to write a very extensive paper as a first draft – on the contrary, it might be more useful to write a short paper, which has the advantage that others will more readily read it and comment on it.

Never be afraid to discuss your study at all stages extensively with others, not only your immediate research colleagues but also semi-outsiders and also in this advance-writing stage. If you know, or are told by others, that a particular direction of your results might not be believed and therefore draw criticism because of some potential deficiency in your study, why not remedy it at this stage? Looking at what you have written, or by discussing potential results with others, you will be able to imagine more clearly what your readers and critical colleagues might object to.

Writing a paper beforehand is the ultimate test of whether the research project is what you wanted, whether your reasoning flows logically, or whether you forgot something. The initial draft will be a yardstick for yourself and for others – whatever happens during the course of your research. This will help you to surmount surprise happenings: you have written down where you started and why, and therefore you will also know very securely when and why you have to take a detour – or even a U-turn.

Writing is difficult and time-consuming. Writing a paper can easily take 5–10 revisions, which might span a full year (inclusive of the time it takes your supervisor or your colleagues to produce comments). During the writing, you will often be obliged to go back to the data and do additional or different analyses. Since your paper will need many revisions, and this will take such a long time, why not take a head-start at the beginning of your data collection? It will save frustration and lost time at the end of your project.

Many guidelines and advices exist about writing, both about the substance (how to use words and phrases) and about the process. All beginning researchers should have a look at some books and papers about writing, and seasoned researchers can still profit from rereading them. Several reporting guidelines exist for several types of studies (RCTs, observational, diagnostic research, etc). They are often very detailed, in describing what should be in title, abstract, and so on. Although they should not be mechanically adhered to, 28 they help writing. In Appendix F , we have collected some wisdom that we particularly liked; several books on writing are listed, as well as reporting guidelines that help researchers to craft papers that are readable and contain all the information that is necessary and useful to others.

Now you can start “your research”

After the piloting and after having written your paper, you are ready to start your data collection, your analysis, or whatever is needed to “do your research”.

The work that is needed before you can start to “do your research” will take a great deal of time and effort. What will you have achieved after setting up a piece of research following the lengthy and involved precepts of this paper? You will have specified a limited research question that you will solve. You will add one little shining stone to the large mosaic of science. At the time that you do the study, you may still be too close to see its effect on the overall picture. That will come over the years.

Further reading

Some texts that we mention in the paper might be especially worthwhile for further reading; see Appendix G .

Acknowledgments

We thank Miguel Hernán, Stuart Pocock, and Bianca De Stavola for their informative comments on an earlier draft manuscript, as well as two anonymous reviewers of Clinical Epidemiology . The Centre for Global NCDs is supported by the Wellcome Trust Institutional Strategic Support Fund (097834/Z/11/B). This work was also supported by the European Research Council under the European Union’s Seventh Framework Programme (FP7/2007-2013 / ERC grant agreement number 668954).

The authors report no conflicts of interest in this work.

Have a language expert improve your writing

Run a free plagiarism check in 10 minutes, generate accurate citations for free.

• Knowledge Base

Methodology

• What Is Qualitative Research? | Methods & Examples

What Is Qualitative Research? | Methods & Examples

Published on June 19, 2020 by Pritha Bhandari . Revised on September 5, 2024.

Qualitative research involves collecting and analyzing non-numerical data (e.g., text, video, or audio) to understand concepts, opinions, or experiences. It can be used to gather in-depth insights into a problem or generate new ideas for research.

Qualitative research is the opposite of quantitative research , which involves collecting and analyzing numerical data for statistical analysis.

Qualitative research is commonly used in the humanities and social sciences, in subjects such as anthropology, sociology, education, health sciences, history, etc.

• How does social media shape body image in teenagers?
• How do children and adults interpret healthy eating in the UK?
• What factors influence employee retention in a large organization?
• How is anxiety experienced around the world?
• How can teachers integrate social issues into science curriculums?

Table of contents

Approaches to qualitative research, qualitative research methods, qualitative data analysis, advantages of qualitative research, disadvantages of qualitative research, other interesting articles, frequently asked questions about qualitative research.

Qualitative research is used to understand how people experience the world. While there are many approaches to qualitative research, they tend to be flexible and focus on retaining rich meaning when interpreting data.

Common approaches include grounded theory, ethnography , action research , phenomenological research, and narrative research. They share some similarities, but emphasize different aims and perspectives.

Note that qualitative research is at risk for certain research biases including the Hawthorne effect , observer bias , recall bias , and social desirability bias . While not always totally avoidable, awareness of potential biases as you collect and analyze your data can prevent them from impacting your work too much.

Prevent plagiarism. Run a free check.

Each of the research approaches involve using one or more data collection methods . These are some of the most common qualitative methods:

• Observations: recording what you have seen, heard, or encountered in detailed field notes.
• Interviews:  personally asking people questions in one-on-one conversations.
• Focus groups: asking questions and generating discussion among a group of people.
• Surveys : distributing questionnaires with open-ended questions.
• Secondary research: collecting existing data in the form of texts, images, audio or video recordings, etc.
• You take field notes with observations and reflect on your own experiences of the company culture.
• You distribute open-ended surveys to employees across all the company’s offices by email to find out if the culture varies across locations.
• You conduct in-depth interviews with employees in your office to learn about their experiences and perspectives in greater detail.

Qualitative researchers often consider themselves “instruments” in research because all observations, interpretations and analyses are filtered through their own personal lens.

For this reason, when writing up your methodology for qualitative research, it’s important to reflect on your approach and to thoroughly explain the choices you made in collecting and analyzing the data.

Qualitative data can take the form of texts, photos, videos and audio. For example, you might be working with interview transcripts, survey responses, fieldnotes, or recordings from natural settings.

Most types of qualitative data analysis share the same five steps:

• Prepare and organize your data. This may mean transcribing interviews or typing up fieldnotes.
• Review and explore your data. Examine the data for patterns or repeated ideas that emerge.
• Develop a data coding system. Based on your initial ideas, establish a set of codes that you can apply to categorize your data.
• Assign codes to the data. For example, in qualitative survey analysis, this may mean going through each participant’s responses and tagging them with codes in a spreadsheet. As you go through your data, you can create new codes to add to your system if necessary.
• Identify recurring themes. Link codes together into cohesive, overarching themes.

There are several specific approaches to analyzing qualitative data. Although these methods share similar processes, they emphasize different concepts.

Qualitative research often tries to preserve the voice and perspective of participants and can be adjusted as new research questions arise. Qualitative research is good for:

• Flexibility

The data collection and analysis process can be adapted as new ideas or patterns emerge. They are not rigidly decided beforehand.

• Natural settings

Data collection occurs in real-world contexts or in naturalistic ways.

• Meaningful insights

Detailed descriptions of people’s experiences, feelings and perceptions can be used in designing, testing or improving systems or products.

• Generation of new ideas

Open-ended responses mean that researchers can uncover novel problems or opportunities that they wouldn’t have thought of otherwise.

Receive feedback on language, structure, and formatting

Professional editors proofread and edit your paper by focusing on:

• Academic style
• Vague sentences
• Style consistency

See an example

Researchers must consider practical and theoretical limitations in analyzing and interpreting their data. Qualitative research suffers from:

• Unreliability

The real-world setting often makes qualitative research unreliable because of uncontrolled factors that affect the data.

• Subjectivity

Due to the researcher’s primary role in analyzing and interpreting data, qualitative research cannot be replicated . The researcher decides what is important and what is irrelevant in data analysis, so interpretations of the same data can vary greatly.

• Limited generalizability

Small samples are often used to gather detailed data about specific contexts. Despite rigorous analysis procedures, it is difficult to draw generalizable conclusions because the data may be biased and unrepresentative of the wider population .

• Labor-intensive

Although software can be used to manage and record large amounts of text, data analysis often has to be checked or performed manually.

If you want to know more about statistics , methodology , or research bias , make sure to check out some of our other articles with explanations and examples.

• Chi square goodness of fit test
• Degrees of freedom
• Null hypothesis
• Discourse analysis
• Control groups
• Mixed methods research
• Non-probability sampling
• Quantitative research
• Inclusion and exclusion criteria

Research bias

• Rosenthal effect
• Implicit bias
• Cognitive bias
• Selection bias
• Negativity bias
• Status quo bias

Quantitative research deals with numbers and statistics, while qualitative research deals with words and meanings.

Quantitative methods allow you to systematically measure variables and test hypotheses . Qualitative methods allow you to explore concepts and experiences in more detail.

There are five common approaches to qualitative research :

• Grounded theory involves collecting data in order to develop new theories.
• Ethnography involves immersing yourself in a group or organization to understand its culture.
• Narrative research involves interpreting stories to understand how people make sense of their experiences and perceptions.
• Phenomenological research involves investigating phenomena through people’s lived experiences.
• Action research links theory and practice in several cycles to drive innovative changes.

Data collection is the systematic process by which observations or measurements are gathered in research. It is used in many different contexts by academics, governments, businesses, and other organizations.

There are various approaches to qualitative data analysis , but they all share five steps in common:

• Prepare and organize your data.
• Review and explore your data.
• Develop a data coding system.
• Assign codes to the data.
• Identify recurring themes.

The specifics of each step depend on the focus of the analysis. Some common approaches include textual analysis , thematic analysis , and discourse analysis .

Cite this Scribbr article

If you want to cite this source, you can copy and paste the citation or click the “Cite this Scribbr article” button to automatically add the citation to our free Citation Generator.

Bhandari, P. (2024, September 05). What Is Qualitative Research? | Methods & Examples. Scribbr. Retrieved September 13, 2024, from https://www.scribbr.com/methodology/qualitative-research/

Is this article helpful?

Pritha Bhandari

Other students also liked, qualitative vs. quantitative research | differences, examples & methods, how to do thematic analysis | step-by-step guide & examples, "i thought ai proofreading was useless but..".

I've been using Scribbr for years now and I know it's a service that won't disappoint. It does a good job spotting mistakes”

How to Choose and Develop a Research Topic: Ideas and Examples

Discover strategies for choosing and developing a compelling research topic. Generate ideas, refine your topic, and conduct effective research.

Kate Windsor

Jun 26, 2024

Selecting the right research paper topic is a crucial step in the research process. A well-chosen topic can lay the foundation for a successful research project, while a poorly chosen one can lead to frustration and wasted effort. Choosing an interesting research topic can be challenging, especially for those new to the research field. 

This article aims to provide guidance and inspiration for researchers seeking to choose and develop a compelling research topic and/or topics to write. 

Understanding the Characteristics of a Good Research Topic

A good research topic should possess several key characteristics:

• Originality and novelty: The topic should contribute new knowledge or insights to the field, rather than simply rehashing existing research.
• **Feasibility and relevance: **The topic should be feasible to research within the given timeframe and resources, and relevant to the researcher's field of study.
• **Significance and impact: **The topic should have the potential to make a significant impact on the field and contribute to the advancement of knowledge.

Strategies for Generating Research Topic Ideas

Generating research topic ideas or thinking of topics to write a research on can be a daunting task, but there are several strategies that can help:

Brainstorming Techniques

• Mind mapping: Create a visual representation of your ideas and how they connect to each other.
• Freewriting: Write down your thoughts and ideas without censoring yourself, and then review what you've written to identify potential topics.
• Questioning: Ask yourself questions about your field of study, such as "What are the current gaps in knowledge?" or "What are the most pressing issues facing the field?".

Exploring Personal Interests and Experiences

Your personal interests and experiences can be a rich source of inspiration for research topics. Consider what you are passionate about and how it intersects with your field of study for your research paper ideas. Choose a topic that interests you.

Keeping Up with Current Trends and Developments

• Reading academic journals and publications: Stay up-to-date with the latest research in your field by regularly reading academic journals and publications.
• Attending conferences and seminars: Attend conferences and seminars to learn about current trends and developments in your field, and to network with other researchers.

Seeking Inspiration from AI for Research

AI for research can be a valuable tool for generating research topic ideas. AI algorithms can analyze vast amounts of data and identify patterns and trends that may not be immediately apparent to human researchers.

Easily pronounces technical words in any field

Narrowing Down and Refining Your Research Topic

Once you have generated some potential research topics, the next step is to narrow down and refine your topic:

• Identifying a broad area of interest: Start by identifying a broad area of interest within your field of study.
• Conducting preliminary research: Conduct preliminary research to gain a better understanding of the existing research in your area of interest.
• Formulating a specific research question: Formulate a specific research question that addresses a gap in the existing research or explores a new angle on a familiar topic. This research question will serve as the basis for your thesis or thesis statement.
• Considering the scope and feasibility of the topic: Consider the scope and feasibility of your topic, taking into account the timeframe and resources available to you.
• Ensuring the topic aligns with the requirements of your research paper or scientific paper: Make sure your topic aligns with the requirements of your research paper or scientific paper, such as word count, formatting, and citation style.

Developing Your Research Topic

Once you have narrowed down and refined your research topic, the next step is to develop it further:

Conducting a Literature Review

• Identifying key sources and references: Identify the key sources and references in your field of study that are relevant to your research topic.
• Synthesizing and analyzing existing research: Synthesize and analyze the existing research to identify gaps in knowledge and potential areas for further exploration.

Formulating Hypotheses or Research Objectives

Formulate hypotheses or research objectives based on your analysis of the existing research and your own insights and observations.

Defining Key Concepts and Variables

Define the key concepts and variables that are central to your research topic, and operationalize them in a way that is measurable and testable.

Outlining the Research Methodology

Outline the research methodology you will use to investigate your research topic, including data collection methods, sampling strategies, and data analysis techniques.

Tips on How to Write Faster and Efficiently

Writing a research paper can be a time-consuming process, but there are several tips and strategies that can help you  write faster and more efficiently:

• Break your writing into manageable chunks and set achievable goals for each writing session.
• Use outlines and mind maps to organize your thoughts and ideas before you start writing.
• Minimize distractions by finding a quiet workspace and turning off notifications on your devices.
• Take regular breaks to recharge and avoid burnout.
• Utilize writing tools and software, such as Grammarly or Scrivener , to streamline your writing process and improve the quality of your work.

Research Topic Ideas and Examples

Here are some examples of research topics in various fields of study:

Social Sciences

• The impact of social media on interpersonal relationships
• The role of education in reducing income inequality

Natural Sciences

• Exploring the potential of renewable energy sources
• Investigating the effects of climate change on biodiversity
• Analyzing the influence of popular culture on literature
• Examining the evolution of language in the digital age

Business and Economics

• The impact of remote work on employee productivity and job satisfaction
• Investigating the role of corporate social responsibility in consumer decision-making

Common Mistakes to Avoid When Choosing a Research Topic

When choosing a research topic, there are several common mistakes to avoid:

• Choosing a topic that is too broad or too narrow : A topic that is too broad may lack focus and depth, while a topic that is too narrow may limit the potential impact and significance of the research.
• Failing to consider the relevance and significance of the topic: A topic that is not relevant or significant to the field may not be worth researching, even if it is personally interesting to the researcher.
• Neglecting to conduct sufficient preliminary research : Failing to conduct sufficient preliminary research can lead to a lack of understanding of the existing research in the field, and may result in a topic that has already been thoroughly explored by other researchers. This can lead to wasted time and effort, as well as a lack of originality in the research.
• Ignoring the importance of a well-crafted research paper title : A well-crafted  research paper title can help to attract readers and convey the significance of the research. Ignoring the importance of the title can lead to a lack of engagement with the research.  A strong title should be concise, informative, and engaging, accurately reflecting the content and purpose of the research.

Choosing and developing a research topic is a crucial step in the research process, and one that requires careful consideration and planning. By understanding the characteristics of a good research topic, employing strategies for generating ideas, narrowing down and refining your topic, and developing it further through a literature review and research methodology, you can set yourself up for success in your research endeavors.

While the process of topic selection can be challenging, it is also an opportunity to explore your passions and interests, and to contribute new knowledge and insights to your field of study. By investing time and effort in selecting a compelling and feasible research topic, you can lay the foundation for a good research paper and a successful and impactful research project. 

Thesis Development

Literature Review

Academic Writing

Research Methodology

Research Idea Generation

Research Topic Selection

Recent articles

What is an Individualized Education Plan (IEP)?

Aug 1, 2024

Individualized Education Plan

Special Education

IEP Process

Learning Disabilities

Assistive Technology

Noam Chomsky's Theory of Language Acquisition

Aug 5, 2024

What are the Responsibilities of a Cosigner in a Student Loan?

Aug 6, 2024

Financial Aid

College Funding

Cosigner Responsibilities

Student Loans

10 Best Productivity Books

Aug 13, 2024

Productivity Books

Time Management

Efficiency Tips

Self Improvement

Goal Setting

Subscribe to the PwC Newsletter

Join the community, edit social preview.

Add a new code entry for this paper

Remove a code repository from this paper, mark the official implementation from paper authors, add a new evaluation result row.

• EXPERIMENTAL DESIGN
• SCIENTIFIC DISCOVERY

Remove a task

Add a method, remove a method, edit datasets, can llms generate novel research ideas a large-scale human study with 100+ nlp researchers.

6 Sep 2024  ·  Chenglei Si , Diyi Yang , Tatsunori Hashimoto · Edit social preview

Recent advancements in large language models (LLMs) have sparked optimism about their potential to accelerate scientific discovery, with a growing number of works proposing research agents that autonomously generate and validate new ideas. Despite this, no evaluations have shown that LLM systems can take the very first step of producing novel, expert-level ideas, let alone perform the entire research process. We address this by establishing an experimental design that evaluates research idea generation while controlling for confounders and performs the first head-to-head comparison between expert NLP researchers and an LLM ideation agent. By recruiting over 100 NLP researchers to write novel ideas and blind reviews of both LLM and human ideas, we obtain the first statistically significant conclusion on current LLM capabilities for research ideation: we find LLM-generated ideas are judged as more novel (p < 0.05) than human expert ideas while being judged slightly weaker on feasibility. Studying our agent baselines closely, we identify open problems in building and evaluating research agents, including failures of LLM self-evaluation and their lack of diversity in generation. Finally, we acknowledge that human judgements of novelty can be difficult, even by experts, and propose an end-to-end study design which recruits researchers to execute these ideas into full projects, enabling us to study whether these novelty and feasibility judgements result in meaningful differences in research outcome.

Code Edit Add Remove Mark official

Tasks edit add remove, datasets edit, results from the paper edit, methods edit add remove.

AIs generate more novel and exciting research ideas than human experts

The first statistically significant results are in: not only can Large Language Model (LLM) AIs generate new expert-level scientific research ideas, but their ideas are more original and exciting than the best of ours – as judged by human experts .

Recent breakthroughs in large language models (LLMs) have excited researchers about the potential to revolutionize scientific discovery, with models like ChatGPT and Anthropic's Claude showing an ability to autonomously generate and validate new research ideas.

This, of course, was one of the many things most people assumed AIs could never take over from humans; the ability to generate new knowledge and make new scientific discoveries, as opposed to stitching together existing knowledge from their training data.

But as with artistic expression, music composition, coding, understanding subtext and body language, and any number of other emergent abilities, today's multimodal AIs do appear to be able to generate novel research – more novel on average than their human counterparts.

No previous research had been done in this field until recently, when over 100 natural language processing (NLP) research experts (PhDs and post-doctorates from 36 different, well-regarded institutions) went head-to-head with LLM-generated 'ideation agents' to see whose research ideas were more original, exciting and feasible – as judged by human experts.

In our new paper: https://t.co/xjhjUC1j8J We recruited 49 expert NLP researchers to write novel ideas on 7 NLP topics. We built an LLM agent to generate research ideas on the same 7 topics. After that, we recruited 79 experts to blindly review all the human and LLM ideas. 2/ pic.twitter.com/GMidcFZgkQ — CLS (@ChengleiSi) September 9, 2024

The field of NLP is a branch of artificial intelligence that deals with communication between humans and AIs, in language that both sides can 'understand,' in terms of basic syntax, but also nuance – and more recently, in terms of verbal tone and emotional inflection .

49 human experts wrote ideas on 7 NLP topics, while an LLM model trained by the researchers generated ideas on the same 7 topics. The study paid US$300 for each idea plus a bonus of $1,000 to the top five human ideas in an effort to incentivize the humans to produce legitimate, easy-to-follow and execute ideas.

Once complete, an LLM was used to standardize the writing styles of each submitted entry while preserving the original content in order to level the playing field, so to speak, keeping the study as blind as possible.

When we say “experts”, we really do mean some of the best people in the field. Coming from 36 different institutions, our participants are mostly PhDs and postdocs. As a proxy metric, our idea writers have a median citation count of 125, and our reviewers have 327. 3/ pic.twitter.com/L2ihAAtJRt — CLS (@ChengleiSi) September 9, 2024

All the submissions were then reviewed by 79 recruited human experts and a blind judgment of all research ideas was made. The panel submitted 298 reviews, giving each idea between two to four independent reviews.

And sure enough, when it comes to novelty and excitement, the AIs tested significantly better than human researchers. They also ranked slightly lower than humans in feasibility, and slightly higher in effectiveness – but neither of these effects were found to be statistically significant.

The study also uncovered certain flaws, such as the LLM's lack of diversity in generating ideas as well as their limitations in self-evaluation. Even with explicit direction not to repeat itself, the LLM would quickly begin to do so. LLMs also weren't able to review and score ideas with much consistency and scored low in agreement with human judgments.

The study also acknowledges that the human side of judging the "originality" of an idea is rather subjective, even with a panel of experts.

To better prove the theory that LLMs may or may not be better at the potential for autonomous scientific discovery, the researchers will recruit more expert participants. They propose a more comprehensive follow-up study, where the ideas generated by both AI and humans are fully developed into projects, allowing for a more in-depth exploration of their impact in real-world scenarios.

But these initial findings are certainly sobering. Humanity finds itself looking a strange new adversary in the eye. Language model AIs are becoming incredibly capable tools – but they're still notoriously unreliable and prone to what AI companies call "hallucinations," and what anyone else might call "BS."

They can move mountains of paperwork – but there's certainly no room for "hallucinations" in the rigor of the scientific method. Science can't build on a foundation of BS. It's already scandalous enough that by some estimates , at least 10% of research papers are currently being co-written – at the very least – by AIs.

On the other hand, we can't understate AI's potential to radically accelerate progress in certain areas – as evidenced by Deepmind's GNoME system , which knocked off about 800 years' worth of materials discovery in a matter of months, and spat out recipes for about 380,000 new inorganic crystals that could have revolutionary potential in all sorts of areas.

This is the fastest-developing technology humanity has ever seen; it's reasonable to expect that many of its flaws will be patched up and painted over within the next few years. Many AI researchers believe we're approaching general superintelligence – the point at which generalist AIs will overtake expert knowledge in more or less all fields.

It's certainly a strange feeling watching our greatest invention rapidly master so many of the things we thought made us special – including the very ability to generate novel ideas. Human ingenuity seems to be painting humans into a corner, as old gods of ever-diminishing gaps.

Still, in the immediate future, we can make the best progress as a symbiosis, with the best of organic and artificial intelligence working together, as long as we can keep our goals in alignment.

But if this is a competition, well, it's AI: 1, humans: 0 for this round.

Source: Chenglei Si via X

Most Viewed

Iconic westfalia camper van streamlined into fast-pitching all-roader, all that glitters is not gold: ola e-motorcycles slammed upon launch, throwable tactical camera transmits 360° panoramic thermal images.

How to Generate Business Ideas with AI Tools

In today’s fast-paced world, coming up with fresh business ideas can be tough. That’s where AI tools come in handy. These tools can quickly analyze data, spot trends, and suggest new ideas, making the brainstorming process much easier and more efficient.

Why AI Tools Are Essential for Generating Business Ideas

AI tools have become a cornerstone in modern entrepreneurship. They help businesses brainstorm faster and think more creatively. With AI, you can go from no inspiration to three unique ideas in just five seconds. This speed is unmatched and can supercharge group discussions and creativity games.

Advantages of Using AI for Idea Generation

Using AI tools for idea generation offers several benefits:

• Speed : AI can generate ideas in seconds, saving valuable time.
• Creativity : AI helps you find unexpected solutions that you might not have considered.
• Scalability : AI tools can handle large amounts of data, making them suitable for businesses of all sizes.

Based on Global State of AI research, Eighty-nine percent of organizations believe AI and machine learning will help them grow revenue, boost operational efficiency, and improve customer experiences.

Steps to Generate Business Ideas Using AI Tools

1.    defining your business goals and interests.

Before diving into AI tools, it’s crucial to define your business goals and interests. This step sets the foundation for the entire idea generation process. Ask yourself what you want to achieve and which areas excite you the most. This will help you focus your efforts and make the most out of AI tools.

2.    Using AI Tools for Initial Brainstorming

Start with using an AI business ideas generator to conduct a brainstorming session. These tools can quickly provide a variety of ideas based on your input. For instance, you can use ChatGPT to cover products, markets, budgets, and regions. Repeat and refine this process until you find an area that excites you.

3.    Leveraging AI for Market Research and Trend Analysis

Once you have some initial ideas, use AI tools for market research and trend analysis. Tools like browse.ai can scrape the web for relevant data, helping you understand market demands and trends. This step ensures that your ideas are not only creative but also viable in the current market.

AI tools can help you find unexpected solutions that you might not have considered otherwise. This can be a game-changer for your business.

Maximizing the Potential of AI-Generated Business Ideas

Once you have a list of ideas from an AI business ideas generator , the next step is to evaluate and refine them. Not all ideas will be perfect right out of the gate. Here are some steps to help you fine-tune your ideas:

• Assess Feasibility: Determine if the idea is practical and can be implemented with the resources you have.
• Market Fit: Check if there is a demand for the idea in the market.
• Scalability: Consider if the idea can grow and scale over time.
• Innovation: Ensure the idea brings something new or improves upon existing solutions.

Aligning Ideas with Business Needs

Aligning AI-generated ideas with your business needs is crucial for success. Here’s how you can do it:

• Identify Core Objectives: Make sure the ideas align with your business goals and objectives.
• Resource Allocation: Ensure you have the necessary resources to implement the idea.
• Stakeholder Buy-In: Get approval and support from key stakeholders to move forward.

Aligning AI-generated ideas with your business needs ensures that you are not just chasing trends but are focused on sustainable growth.

Popular AI Tools for Business Idea Generation

AI tools have become essential for generating business ideas across various sectors like technology, health care, and transportation. These tools can help businesses, both big and small, as well as individual consumers, to brainstorm faster and think more creatively.

Features to Look for in AI Tools

When choosing an AI tool for generating business ideas, consider the following features:

• Ease of Use: The tool should be user-friendly and require minimal setup.
• Customization: Look for tools that allow you to adjust settings to fit your specific needs.
• Security: Ensure the tool uses SSL encryption to keep your ideas secure.
• Integration: The tool should integrate well with other platforms you use, like marketing plan makers or marketing strategy generators.

In today’s fast-paced world, coming up with new business ideas can be tough. But with AI tools , this process becomes much easier and quicker. These tools can help you spot trends, understand market needs, and even come up with unique ideas you might not have thought of on your own. So, why not give it a try?  Use Toolio.ai which is a free tools directory . Embrace the power of AI and watch your business ideas grow!

Lakisha Davis is a tech enthusiast with a passion for innovation and digital transformation. With her extensive knowledge in software development and a keen interest in emerging tech trends, Lakisha strives to make technology accessible and understandable to everyone.

Type above and press Enter to search. Press Esc to cancel.

Can Large Language Models Unlock Novel Scientific Research Ideas?

“An idea is nothing more nor less than a new combination of old elements" Young ( 2019 ) . The widespread adoption of Large Language Models (LLMs) and publicly available ChatGPT have marked a significant turning point in the integration of Artificial Intelligence (AI) into people’s everyday lives. This study explores the capability of LLMs in generating novel research ideas based on information from research papers. We conduct a thorough examination of 4 LLMs in five domains (e.g., Chemistry, Computer, Economics, Medical, and Physics). We found that the future research ideas generated by Claude-2 and GPT-4 are more aligned with the author’s perspective than GPT-3.5 and Gemini. We also found that Claude-2 generates more diverse future research ideas than GPT-4, GPT-3.5, and Gemini 1.0. We further performed a human evaluation of the novelty, relevancy, and feasibility of the generated future research ideas. This investigation offers insights into the evolving role of LLMs in idea generation, highlighting both its capability and limitations. Our work contributes to the ongoing efforts in evaluating and utilizing language models for generating future research ideas. We make our datasets and codes publicly available 1 1 1 https://github.com/sandeep82945/Future-Idea-Generation .

[ topline=True, bottomline=True, skipabove=skipbelow=leftmargin=0cm, rightmargin=0cm, innertopmargin=10pt, innerbottommargin=10pt, innerleftmargin=0pt, innerrightmargin=3pt, linewidth=2pt, linecolor=blue, backgroundcolor=gray!20 ]mybox

Sandeep Kumar † † \dagger † , Tirthankar Ghosal ‡ ‡ \ddagger ‡ , Vinayak Goyal † † \dagger † , Asif Ekbal † † \dagger † † † \dagger † Indian Institute of Technology Patna, India ‡ ‡ \ddagger ‡ National Center for Computational Sciences, Oak Ridge National Laboratory, USA † † \dagger † (sandeep_2121cs29,2201ai52_vinayak,asif)@iitp.ac.in ‡ ‡ \ddagger ‡ [email protected]

“Innovation is seeing what everybody has seen and thinking what nobody has thought” —Dr. Albert Szent-Györgyi

1 Introduction

An idea can be defined as a thought or suggestion aimed at solving a problem or considering a possibility. This concept is central to fields ranging from philosophy to science and economics. According to Plato et al. ( 2000 ) , ideas are archetypal forms that represent the most accurate reality. In the context of scientific research, Kuhn and Hawkins ( 1963 ) in "The Structure of Scientific Revolutions" describes an idea as a realization or hypothesis that can challenge and shift paradigms within a scientific community. Therefore, an idea can be understood as a cognitive construct that arises from the human mind’s ability to process information, reflect, and imagine, serving as a cornerstone for creativity, problem-solving, and innovation. Idea generation can be generally understood as a state of focused internally-directed attention involving controlled semantic retrieval Benedek et al. ( 2014 ) .

As technology improves, new capabilities emerge. Ever since the Turing Test was proposed in the 1950s, humans have explored the mastering of language intelligence by machine Zhao et al. ( 2023 ) . Technological advancements serve two key functions in innovation. Firstly, they influence the goals of generating and selecting ideas. Secondly, they impact the methodology of how ideas are generated and chosen Kornish and Hutchison-Krupat ( 2017 ) . LLMs have exhibited unparalleled mastery of natural language processing (NLP). Since, these have become increasingly powerful, researchers have begun to investigate their reasoning ability in problem-solving tasks Yao et al. ( 2022 ); Brahman et al. ( 2023 ) . The concept of an idea is essentially a new combination of old elements. LLMs have access to a broad spectrum of knowledge, due to their extensive training on vast amounts of text data. However, understanding how information extracted from a research paper can give rise to new ideas, which have not yet been explored much. This leads us to ponder:

Motivated by this, in this paper, we analyze the potential of LLMs in generating future research directions/ideas. As LLMs possess knowledge across various domains, we investigate five specific areas, viz. Computer Science, Physics, Chemistry, Economics, and Medicine. To address this task, we create a dataset of papers published after the year 2022 from these five domains. We annotate the papers with future research ideas. To evaluate the novelty and relevance of ideas generated by the LLMs, we propose an Idea Alignment Score (IAScore). This score reflects how well the generated ideas align with those proposed by the authors. To study the model’s ability to generate diverse ideas, we propose an Idea Distinctness Index. We analyze and discuss the performance and limitations of four LLMs: Gemini Anil et al. ( 2023 ) , Claude-2 Anthropic ( 2023 ) , GPT-3.5, and GPT-4 OpenAI ( 2023 ) . We further conduct a human evaluation of 460 generated ideas in computer science to study the novelty, relevance, and feasibility of these research ideas. This paper demonstrates that LLMs have the potential to generate relevant, distinct, feasible, and novel ideas to some extent.

To summarize, our main contributions in this paper are:

We contribute to the ongoing exploration of LLMs’ capabilities in generating future research ideas.

To address the task, we create a novel dataset of recent papers of five domains (Computer science, Economics, Chemistry, Physics, Medical).

To access the quality of generated ideas from LLMs, we propose Idea Alignment Score and Idea Distinctness Index to evaluate the idea generation capability.

We discuss the challenges associated with human evaluation and conduct a human evaluation on 460 generated ideas.

We hope that this work serves as a foundation for future studies focused on accelerating scientific research by automatically generating research ideas.

2 Related Work

Recently, LLMs have shown emergent abilities to perform tasks they were not explicitly trained for Wei et al. ( 2022 ); Bubeck et al. ( 2023 ) . This includes common sense question answering, code generation, and cross-domain problem solving, enriching their utility across unforeseen domains Chen et al. ( 2021 ); Sarsa et al. ( 2022 ) . Their capability extends to advanced scientific domains such as computer science, physics, medicine, and mathematics Romera-Paredes et al. ( 2023 ); Huang et al. ( 2023 ) . Technology Semantic Network (TechNet) was proposed to stimulate idea generation in engineering design Sarica et al. ( 2021 ) . There have been a few works in the discovery of new proteins to accelerate scientific discovery. The prior work reported in Spangler et al. ( 2014 ) involves utilizing published studies to find new protein kinases that phosphorylate the tumor suppressor protein p53.

A hypothesis is a hunch, assumption, suspicion, assertion or an idea about a phenomenon, relationship or situation, the reality or truth of which you do not know Kumar ( 1996 ) . There have been some works on hypothesis generation. Initial studies on automated hypothesis generation begin by constructing a corpus of distinct concepts. Subsequently, they explore the relationships between these concepts using machine learning techniques, such as analyzing the similarities among vectors representing different words (or concepts) Tshitoyan et al. ( 2019 ) , or applying link prediction methods over a graph (where concepts are nodes) Nadkarni et al. ( 2021 ) . Recently Qi et al. ( 2023 ) used LLMs and extensive pre-existing knowledge of various scientific fields for hypothesis generation. PaperRobot Wang et al. ( 2019 ) predicts related entities for an input title and writes key elements of a new paper, including the abstract, conclusion, and future work, and predicts a new title.

Xu et al. ( 2023 ) developed a framework that leverages the concept co-occurrence graphs and a masked language model to explore and verbalize academic ideas. Their method involves constructing evolving concept graphs across various disciplines and utilizing temporal link prediction to identify potential interdisciplinary connections. The framework also incorporates pre-trained language models to articulate these connections in a coherent academic context. SciMON Wang et al. ( 2023 ) showed that LLMs can be guided by seed terms to generate specific ideas. They applied a pre-trained sentence classifier to classify sentences from the title and abstract into categories of Background, Method, Objective. They considered sentences labeled as problems or motivations as background, and the remaining were treated as target output sentences. Additionally, a pre-trained entity extractor was used to extract salient seed terms.

However, previous works primarily focused on developing methods (linking and explaining entities, which may not sufficiently capture the complexity or explain how LLMs can solve real-world problems) for idea generation, whereas our work exhaustively focuses on evaluating the capability of LLMs in generating research ideas. Our goal is to assess the inherent ability of LLMs to generate future research ideas/directions.

Our dataset creation involves three steps: (1) Dataset Collection, (2) FRI Identification and removal, and (3) FRI generation.

3.1 Dataset Collection

We construct a corpus D from S2ORC collected 100 papers from the domains of Computer Science, Economics, Physics, Chemistry, Medical from Lo et al. ( 2020 ) . To ensure the quality and relevance of the data and to utilize the future research ideas mentioned in a paper, the selected papers must meet the following requirements: (1) the paper must contain the full content, and (2) the paper must include a section on future work.

3.2 FRI Identification and Removal

We first identify and remove any potential research ideas mentioned in the paper. By doing this, we ensure that the LLMs have no prior access to these ideas, which could otherwise affect the objectivity of the analysis.

3.2.1 Annotation Guidelines

Inspired by Hao et al. ( 2020 ) , we define a future research idea as a discussion that the authors believe they will conduct in the future or believe needs to be investigated in future research. We discuss more details about the annotation guidelines in Appendix A .

3.2.2 Annotator Training

Given the complexity of the papers and their frequent use of technical terminology, we hired two doctoral students, each boasting over four years of experience in scientific research publishing. To facilitate their training, an expert with more than ten years of experience in scientific publishing annotated 20 random papers from each domain, adhering to our guidelines. After this initial round of annotation, we reviewed and corrected any misinterpretations with the annotators, further refining their training and enhancing the clarity of our annotation guidelines. To assess the effectiveness of the initial training, we compiled another 20 papers from each domain. From the second round onwards, the annotators demonstrated improved proficiency, accurately identifying at least 95% of the future research ideas on average.

We discuss more details about the annotation process and annotator’s pay in Appendix A .

3.2.3 Future Work Removal

We observed two types of future research ideas (FRIs) 2 2 2 In this paper, we use the terms ‘ideas,’ ‘research ideas,’ ‘future research ideas,’ and ‘FRI’ interchangeably to frequently refer to future research ideas. (Direct FRI and Mixed FRI). We discuss them in details in Appendix H .

AP-FRI Corpus: We removed the sentence from the paper’s input text if it pertains to Direct FRI. However, in the case of Mixed FRI, we did not entirely remove the sentences; instead, we eliminated only parts of sentences or markers indicating future research ideas. We added the removed future ideas to a corpus, which we refer to as the AP-FRI (Author Perspective Future Research Idea Corpus). This corpus contains the future research ideas proposed by the authors of the paper. Also, before adding to the AP-FRI corpus, we merged the sentences about the same topic into a single group.

3.3 FRI Generation using LLM

We investigate various prompts and utilize the following prompts to generate FRIs for papers.

Here, ‘[paper text]’ contains the full content of the paper after removal of future work sections.

3.4 Data Statistics

Figure 3 provides a domain-wise distribution of the average word count in academic papers, excluding discussions on future work (FWK). It can be observed that the length of papers across all fields falls within a range of 7,000 to 8,000 words. Additionally, we calculated the average word count of extracted future work within each domain, providing comparative insights into how different fields prioritize discussions of future research directions. Figure 4 compares the average word count of future work text across six distinct scholarly domains. We observed that the literature in Computer Science notably prioritizes extensive discourse on future research, with an average word count significantly higher than that of other disciplines. In contrast, the literature in Chemistry demonstrates a more concise approach to discussions of future research, as evidenced by its lower average word count.

4 Experiments

4.1 challenges.

To accurately assess the novelty, relevance, and applicability of ideas generated by LLMs, evaluators must possess a high level of expertise in the specific domain and a deep understanding of the research topic to fully grasp the context. Additionally, they need knowledge of related literature to evaluate the ideas’ future potential and the broader implications of their implementation.

4.2 Idea Alignment Score (IAScore)

With the above challenges, the evaluation of ideas generated by LLMs is a challenging process that demands a high number of domain-specific experts. We, therefore, proposed an Idea Alignment Score ( IAScore ), which reflects how well the generated ideas align with those proposed by the author. The underlying idea for this score is that authors of accepted papers can be regarded as experts in their respective subjects. The reason being that they possess thorough background knowledge and have conducted deep analyses of the research topic before getting the paper accepted. Consequently, they are well-acquainted with the pertinent challenges which also may have been discussed by expert reviewers.Therefore, we propose that future ideas mentioned by the authors in the paper could be utilized as good quality of potential FRIs.

The IAScore quantifies the alignment of newly generated ideas with author’s perspectives within a specific domain, and is computed via a two-step process, detailed in Equations 1 and 2 .

Initially, we compute the average alignment score AvgScore j subscript AvgScore 𝑗 \text{AvgScore}_{j} AvgScore start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT for each paper’s ideas. The IdeaMatcher model measures the alignment between the paper’s author Future Research Ideas ( AP-FRI j subscript AP-FRI 𝑗 \text{AP-FRI}_{j} AP-FRI start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT ) and its each generated idea I i ⁢ j subscript I 𝑖 𝑗 \text{I}_{ij} I start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT . The subscript i 𝑖 i italic_i indexes the i 𝑖 i italic_i -th idea within the j 𝑗 j italic_j -th paper, where N j subscript 𝑁 𝑗 N_{j} italic_N start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT represents the total number of ideas proposed in that paper.

"Here, we refer to IM as ’IdeaMatcher’.

LLMs may generate new ideas that even the author may not have thought of. They can also generate additional future ideas, which may or may not be useful. Our goal is for this score is that the LLMs must have generated at least the author’s proposed potential future ideas. Therefore, in our formula of AvgScore j subscript AvgScore 𝑗 \text{AvgScore}_{j} AvgScore start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT , the sum of the alignment scores for a paper’s ideas is divided by the total number of the author’s proposed ideas, N j subscript 𝑁 𝑗 N_{j} italic_N start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT , to normalize the score.

Subsequently, we aggregate the individual paper scores to calculate the domain-wise IAScore . This aggregation, presented in Equation 2 , averages the AvgScore j subscript AvgScore 𝑗 \text{AvgScore}_{j} AvgScore start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT values across all P 𝑃 P italic_P papers within the domain. Higher the value of IAScore domain subscript IAScore domain \text{IAScore}_{\text{domain}} IAScore start_POSTSUBSCRIPT domain end_POSTSUBSCRIPT signifies the more alignment of the generated ideas with author’s perspective of all papers generated by model M 𝑀 M italic_M .

4.2.1 IdeaMatcher

To select an effective IdeaMatcher, we create a small annotated corpus. Our dataset was divided using the standard 30:70 ratio for validation and test sets, respectively. Since our study involves comparing two ideas using a pre-trained model, we did not require a separate training set. We first manually searched for matching pairs of ideas from generated ideas and AP-FRI of the paper. After obtaining 61 matching pairs, we searched for non-matching pairs of ideas, which is straightforward as only one generated idea will match or would not match with another one from AP-FRI while others would not match, so we picked an equal number of non-matching pairs. Then, we experimented with the idea-matching task by considering it similar to the Natural Language Inference (NLI) task. In particular, we considered the generated FRIs to be hypotheses and their corresponding AP-FRIs of the paper to be premises. If the idea matches, the hypothesis should be entailed by the premise. In particular, we used a pre-trained RoBERTa MNLI model Liu et al. ( 2019 ) for this task. We found that this technique produces many false negative cases, resulting in an accuracy of 65.5%.

We also evaluated the idea-matching capability of BERTScore Zhang* et al. ( 2020 ) , as it utilizes BERT embeddings for comparison. We discuss the details in Appendix F . We found that BERTScore performed better than the entailment technique, resulting in an accuracy of 75.4%. We also tried GPT by prompting it with various questions and found that it resulted in 91.8% accuracy when prompted with a specific question prompt below:-

We found that GPT performs better than the existing NLI (Natural Language Inference) and similarity measure such as BERTScore. Therefore, we chose GPT for this task 3 3 3 We used the OpenAI model GPT-3.5-turbo-0125 using OpenAI API .

4.3 Idea Distinctness Index

Distinct-N Li et al. ( 2015 ) , is a metric that measures the diversity of a sentence. It focuses on the number of distinct n-grams of a sentence, and thus penalizes sentences with a lot of repeated words. However, comparing two ideas need semantic comparisons rather than just syntactic differences. So, we introduce a method to semantically evaluate the distinctness of the generated ideas. This method in particular leverages semantic embedding to capture the essence of each idea and computes their distinctness based on semantic similarity measures.

Given a set of generated ideas I = { id 1 , id 2 , … , id n } 𝐼 subscript id 1 subscript id 2 … subscript id 𝑛 I=\{\text{id}_{1},\text{id}_{2},\ldots,\text{id}_{n}\} italic_I = { id start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , id start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , … , id start_POSTSUBSCRIPT italic_n end_POSTSUBSCRIPT } , representing individual ideas, we first encode each idea into a high-dimensional vector space using a pre-trained BERT model Devlin et al. ( 2019 ) 4 4 4 bert-base-uncased 𝐁𝐄𝐑𝐓 : i ⁢ d i ↦ 𝐯 i : 𝐁𝐄𝐑𝐓 maps-to 𝑖 subscript 𝑑 𝑖 subscript 𝐯 𝑖 \mathbf{BERT}:id_{i}\mapsto\mathbf{v}_{i} bold_BERT : italic_i italic_d start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ↦ bold_v start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT , where 𝐯 i ∈ ℝ d subscript 𝐯 𝑖 superscript ℝ 𝑑 \mathbf{v}_{i}\in\mathbb{R}^{d} bold_v start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ∈ blackboard_R start_POSTSUPERSCRIPT italic_d end_POSTSUPERSCRIPT is the embedding of idea i ⁢ d i 𝑖 subscript 𝑑 𝑖 id_{i} italic_i italic_d start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT and d 𝑑 d italic_d is the dimensionality of the embedding space.

To quantify the distinctness between pairs of ideas, we compute the cosine similarity between their embeddings, s ⁢ i ⁢ m ⁢ ( 𝐯 i , 𝐯 j ) = 𝐯 i ⋅ 𝐯 j ‖ 𝐯 i ‖ ⁢ ‖ 𝐯 j ‖ 𝑠 𝑖 𝑚 subscript 𝐯 𝑖 subscript 𝐯 𝑗 ⋅ subscript 𝐯 𝑖 subscript 𝐯 𝑗 norm subscript 𝐯 𝑖 norm subscript 𝐯 𝑗 sim(\mathbf{v}_{i},\mathbf{v}_{j})=\frac{\mathbf{v}_{i}\cdot\mathbf{v}_{j}}{\|% \mathbf{v}_{i}\|\|\mathbf{v}_{j}\|} italic_s italic_i italic_m ( bold_v start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT , bold_v start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT ) = divide start_ARG bold_v start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ⋅ bold_v start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT end_ARG start_ARG ∥ bold_v start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ∥ ∥ bold_v start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT ∥ end_ARG , for each pair of ideas ( i ⁢ d i , i ⁢ d j ) 𝑖 subscript 𝑑 𝑖 𝑖 subscript 𝑑 𝑗 (id_{i},id_{j}) ( italic_i italic_d start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT , italic_i italic_d start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT ) in I 𝐼 I italic_I . The distinctness D i ⁢ j subscript 𝐷 𝑖 𝑗 D_{ij} italic_D start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT between two ideas i 𝑖 i italic_i and j 𝑗 j italic_j is then inversely related to their similarity: D i ⁢ j = 1 − s ⁢ i ⁢ m ⁢ ( 𝐯 i , 𝐯 j ) subscript 𝐷 𝑖 𝑗 1 𝑠 𝑖 𝑚 subscript 𝐯 𝑖 subscript 𝐯 𝑗 D_{ij}=1-sim(\mathbf{v}_{i},\mathbf{v}_{j}) italic_D start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT = 1 - italic_s italic_i italic_m ( bold_v start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT , bold_v start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT ) .

The overall distinctness of the set I 𝐼 I italic_I is calculated as the mean of all pairwise distinctness scores:

This measure provides a single scalar value D I subscript 𝐷 𝐼 D_{I} italic_D start_POSTSUBSCRIPT italic_I end_POSTSUBSCRIPT that quantifies the average diverseness of ideas within a corpus of ideas, with higher values indicating a greater degree of diverseness among the ideas.

Subsequently, we aggregated the distinctness scores across all ideas in each paper to compute the mean distinctness for that paper. Let P = { p 1 , p 2 , … , p m } 𝑃 subscript 𝑝 1 subscript 𝑝 2 … subscript 𝑝 𝑚 P=\{p_{1},p_{2},\ldots,p_{m}\} italic_P = { italic_p start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_p start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , … , italic_p start_POSTSUBSCRIPT italic_m end_POSTSUBSCRIPT } represent the set of papers in a domain, where m 𝑚 m italic_m is the number of papers in the domain. Finally, for a comprehensive assessment of model performance within a domain, we averaged the mean distinctness scores of all papers generated by model M 𝑀 M italic_M as follows:

4.4 Human Evaluation

The evaluation of generated future ideas necessitates familiarity with both previous works related to the subject and the work being evaluated. Specifically, the evaluator must be an expert in the domain and topic. Given the complexity of human evaluation, we approached authors (as the authors have the knowledge of their paper and they also have knowledge of the literate) who have published papers in reputable venues, possess over 5 years of experience in scientific publishing, and have authored more than 5 scientific papers. We collected their accepted papers ( published within 2023 and 2024) and followed the dataset preparation as we discussed in Section 3 and generated FRIs. We modify the prompt slightly to specifically generate only the top five results (see Appendix B ). We selected the outputs from Claude and GPT-4 5 5 5 We used gpt-4-turbo using OpenAI API for the generation models due to their better IAScore and Idea Distinction index. We adopt this approach to avoid author exhaustion and to get an accurate evaluation. We ask the following questions from each human evaluator:-

Q1: Is the idea relevant with the research topic of the paper. (Relevant/Not relevant)

Q2: Assess the originality/novelty of the research idea (5 scale)

Q3: Review the research idea for factual correctness and feasibility. Is the idea impractical or too vague to be actionable? (Not Possible/Possible)

For Q2, we used Best-Worst Scaling Louviere et al. ( 2015 ) on a 5-point scale.

More details about the human evaluation are mentioned in the Appendix B .

5 Results and Discussion

5.1 alignment results.

Figure 5 provides a comparative overview of the IAScore for four language models 6 6 6 We set maximum token length to 512, and temperature=0 for each models Claude-2, Gemini-1.0, GPT-3, and GPT-4 across five academic domains: Chemistry, Computer Science, Economics, Medical, and Physics.

In the Chemistry and Economics domains, Claude has the highest IAScore, indicating strong alignment with the authors’ future research ideas. Claude and GPT-4 have almost similar values for the Computer, Medical, and Physics domains (with GPT-4 slightly higher). GPT-3 and Gemini have lower scores than both GPT-4 and Claude in every domain. GPT-3 has almost the same score as Gemini in the Chemistry and Economics domains. However, it scores higher than Gemini in the Computer, Medical, and Physics domains. The results underscore the advancements in language model capabilities, with each model showcasing domain-specific strengths in idea generation. This alignment of LLMs shows that LLMs are able to generate relevant and novel ideas to some extent. We also studied the effect of length of future work on IAScore (See Appendix D ). We also conducted a human analysis to understand the quality of research ideas generated when the IAScore is low (see Appendix G ).

5.2 Distinctness Results

We show the comparative evaluation of idea distinctness scores in Figure 6 . The line graph depicts the variation of distinctness between the generated ideas and the human-written ideas (AP-FRIs). GPT-3 shows the least distinctness among the generated ideas, except in the Computer domain, where it is slightly more distinct than Gemini. As shown in the graph, the distinctness of Gemini is also quite low; however, it is slightly better than GPT-3, except in the Computer domain.

The generated ideas of GPT-4 are more distinct than those of Gemini and GPT-3 (except for economics, whereas the distinctness of GPT-4 is the same as Gemini). However, it is lower than both Claude and Human. The Idea Distinctness Index of the generated ideas from Claude are almost the same as those of humans for Chemistry, Economics, and Medical domains. However, they are higher than even human scores in the Computer and Physics domains, which shows that it generates very distinct FRIs.

5.3 Human Evaluation Results

We conducted a human evaluation on 460 generated ideas for 46 papers in the computer science domain. To validate the quality of human annotation, we measure the inter-annotator agreement ratio where 20% of the generated ideas are evaluated by two different authors of the same paper. We measured Cohen’s kappa coefficient Cohen ( 1960 ) , which was 0.83, thereby confirming the high quality of the annotations of generated research ideas.

Novelty: Figure 7 displays the results of the human evaluation. We observed that Claude generates 14.78% of non-novel and 16.52% generic FRIs, 41.73% moderately novel, 20.86% very novel, and 16.52% extremely novel FRIs. GPT generates 7.83% not-novel, 13.91% generic, 42.61% moderately novel, 28.70% very novel, and 6.96% extremely novel ideas. Claude generates more non-novel and generic ideas than GPT-4, while GPT-4 produces more very novel ideas and nearly the same number of excellent ideas. This demonstrates that although LLMs also generate generic or already explored ideas, they are capable of producing novel ideas that have either not been explored or have been minimally explored.

Relevance and Feasibility: After human evaluation, we found that that 76.67% of the ideas generated by Claude and 93.34% by GPT-4 are relevant. Furthermore, 83.34% of Claude’s generated ideas and 96.64% of GPT-4’s ideas were judged to be practically feasible and factually correct. These results highlight that Claude and GPT-4 can generate relevant and feasible research ideas. However, the reason Claude generates more impractical and irrelevant research ideas may be that Claude attempts to generate more distinct research ideas than GPT-4, as we evaluated and discussed in Section 5.2 .

5.4 Open-ended generation:

We tested whether LLMs could retain open-ended generation capabilities by providing only a title and abstract as input. Our findings showed that, overall, LLMs can still generate open-ended content due to their past knowledge. However, they may not produce many high-quality ideas, as they lack access to recent publications and methodological insights relevant to the current paper. We discuss this in detail in Appendix C .

5.5 Adding additional background knowledge

We designed our framework based on the Retrieval-Augmented Generation (RAG) model Lewis et al. ( 2020 ) to integrate background knowledge into LLMs, as illustrated in Figure 9 . We collected the titles and abstracts of around 1.9 lakh computer science research papers. Using BERT embeddings, we created vector representations of these titles and stored them in a vector database. From there, we retrieved the 20 papers most similar to our target paper’s title. We extracted contributions from these papers’ abstracts to gather relevant data and then generated ideas by prompting GPT-4 with the target paper and the retrieved background knowledge. We found that adding background knowledge reduced the generation of generic or non-novel ideas and improved relevance and factual accuracy. However, further research is needed to boost the novelty of generated ideas. We discuss this in detail in Appendix E .

6 Conclusion and Future Work

In conclusion, we present the first attempt to evaluate the potential of LLMs in generating future research ideas across five domains: Computer Science, Economics, Chemistry, Physics, and Medicine. Our results and analysis show that LLMs possess domain-specific strengths in idea generation. Furthermore, the results from the Idea Distinctness Index indicate that LLMs, such as Claude and GPT-4, generate distinct research ideas than Gemini and GPT 3.5. GPT-4 and Claude aligns better with authors written future research ideas than Gemini and GPT-4. The alignment of LLMs with the authors of generated ideas, and our human evaluations on relevance, novelty, and feasibility, reveal that although LLMs often produce non-novel and generic ideas, they have the potential to generate relevant and novel and diverse ideas to a significant extent. We hope that the findings and experiments of this work will unlock the potential of LLMs in idea generation and will foster new advancements in automated scientific innovation.

In future work, we plan to investigate more effective way of integrating knowledge from multiple papers to enhance the novelty of ideas generated and prevent the generation of generic and existing ideas.

7 Limitations

7.1 limitations of data collection.

We extracted papers using the Semantic Scholar Academic Graph API from January 2023 to February 2024. The number of papers available is limited by the scope of our data extraction from the Semantic Scholar Academic Graph. We excluded papers that are not in English, as well as those whose abstracts could not be correctly parsed from the PDFs. Not all of these papers include sections on future work; therefore, we annotated only those that contained sections outlining future research directions. So due to such limitations, we collected 100 papers from each domain for analysis.

7.2 Memorization

Carlini et al. ( 2022 ) highlight that LLMs are prone to memorizing portions of their training data, a significant concern in the evaluation of contemporary LLMs. Despite this, the data used for pre-training and post-training includes "a small amount" of more recent data. Therefore, we gathered recent papers from 2023 and 2024. By focusing our evaluation on papers published in these years, the likelihood of test papers appearing in the pre-training corpora for the models is substantially reduced. In addition, we conducted a manual review of these papers to assess memorization. This involved asking various questions related to the papers, such as their titles, publishing venues, author names, etc., to see if the models could supply the missing information. Our findings showed no evidence of such memorization occurring. A similar approach is also followed by Wang et al. ( 2023 ) (discussed in Section 6.4) and even they did not find any evidence of this occurring.

Ethics Statement

We have utilized the open source dataset for our work. Our aim for this work is to assess the potential of language models in generating ideas. Our Institutional Review Board (IRB) evaluated and approved this study. We do not encourage the use of LLMs to generate AI generated research papers (by generating new ideas) or misuse it for harmful idea generation. LLMs can process and synthesize vast amount of literature faster than humans, potentially identifying new patterns or gaps in research that might not be obvious, thus accelerating scientific discovery. However, since LLMs can generate content that may be similar to existing materials, this raises concerns about intellectual property rights and the originality of ideas. LLMs utilized for generating ideas might be misapplied to produce harmful materials such as plans for schemes for designs for destructive devices, explosive devices, ideas for spamming. Notably, it is a common challenge among existing LLMs with strong creative and reasoning abilities. So, we emphasize the responsible use of LLMs for idea generation and the need to broadly improve the safety of LLMs.

Frequently Asked Questions (FAQs)

How does our work differ from Scimon? ⇒ ⇒ \Rightarrow ⇒ Our paper is fundamentally different from the Scimon paper. We would like to highlight a few major differences. While the focus of Scimon is on developing a framework that generates novel scientific ideas, we clarify that our focus is not on generating ideas but on evaluating the capability of LLMs to generate future research ideas/works. We proposed the novel Idea Alignment Score (IAScore) and the Idea Distinctness Index. Unlike Scimon, we approached authors who are knowledgeable about their paper topics and the broader literature (see Section 4.4). Scimon used only GPT for comparison, while we used GPT-4, GPT-3.5, Claude, and Gemini models. Unlike Scimon, we provide the full paper as input. Scimon used the proposed idea written in the abstract as the target, while we used the future work section written in the full paper as our target. Additionally, they utilized a classifier for this purpose, whereas we employed human evaluators, resulting in fewer chances of error and better evaluation results. Our findings are completely different from those of Scimon. We created a novel annotated dataset for these experiments. While Scimon only experimented with computer science papers from the ACL Anthology, we expanded our experiments to five different domains. Scimon generated ideas guided by seed terms to generate specific ideas. Nonetheless, our goal here is to assess the inherent ability of LLMs to generate future work independently. Introducing external aids or additional context would shift the focus from evaluating the LLM’s standalone capabilities to assessing its performance under enhanced conditions. Such an approach would not align with our objective, which is to understand and measure the raw, unaided generative power of LLMs.

Does incorporating extra contextual information alongside individual papers prove counterproductive? ⇒ ⇒ \Rightarrow ⇒ A paper encompasses not only its contributions, findings, and methodology, but also includes the related work and introduction sections, which contain significant background information. It is likely that the major recent related papers pertinent to the current work have already been mentioned. Additionally, LLMs possess general knowledge about the many older papers and the paper itself contains some of the most important related papers. However, we also conducted an experiment to understand the effect of adding additional information (using the RAG framework). We discuss the results and details in Appendix E of the paper. Overall, we observed that incorporating additional background knowledge can somewhat help prevent the generation of non-novel or generic ideas. However, further research is needed to enhance the ability of LLMs to generate more novel ideas.

Appendix A Dataset Annotation

A.1 dataset annotation guidelines.

Recognizing future research idea in a paper involves analyzing the portion of text containing directions for future research. The following steps can be followed:

Step 1: Begin by reading the Title and Abstract of the paper to gain an understanding of its subject matter. It is important to read these sections multiple times to grasp the paper’s main points, such as its motivation, contributions, and other relevant aspects. If necessary, refer to the paper itself or read related material to enhance your understanding.

Step 2: Identify Key Sections for Analysis Focus primarily on the Discussion and Conclusion sections of the paper, as these areas often contain explicit mentions of future research directions. Scan the Methodology section as well, as sometimes suggestions for improving future studies or addressing current study limitations are mentioned here.

Step 3: Distinguish Future Research Ideas from General Statements: Differentiate explicit future research suggestions from general discussion. Future research directions usually involve specific recommendations, plans, or identified gaps that require further exploration. These are often phrased using terms like "future studies should," "further research is needed," or "additional work will." Avoid confusing these with broader statements of potential relevance or applicability, which do not provide direct guidance on future work.

We offer multiple examples of papers with its future research ideas to assist and direct the annotators. We found a few text which looks like future work but is on contrary the motivation of the work. As an example, consider the following: "The goal of this work was to direct attention to emerging and novel research involving "magnetogel nanohybrid materials" that might be relevant in future applications for the treatment of wastewater, as well as in other fields.

The second example is: "Our data could be useful for designing high-quality trials in the future to define the exact role of hemoadsorption in ARDS." . Here, how novel research involving magnetogel nanohybrid material will help in future application is written.

Also another example is: "The goal of this work was to direct attention to emerging and novel research involving magnetogel nanohybrid materials that might be relevant in future applications for the treatment of wastewater, as well as in other fields." This is the application in future, and not the future work.

Step 4: Separate Future Research from Limitations: Carefully examine any limitations mentioned in the paper to determine if they are explicitly linked to future research. Only consider a limitation as future work if the authors clearly indicate a direct intention to address it in subsequent studies. This helps avoid assuming that all limitations naturally lead to future research directions.

There is also very thin line between limitation and future work, where a limitation can or cannot be a future work. There were few cases where limitations were mentioned "One limitation of this paper is the absence of a coordinated attention structure to capture cross-channel information." . As limitations can or cannot be a future work, we only take those limitations which is explicitly mentioned by the author to be a future work. Hence, we only considered the explicit mention of the future work by the author in their paper.

A.1.1 Annotator Training

A.1.2 annotation process.

We regularly monitored the annotated data, placing emphasis on identifying and rectifying inconsistencies and cases of confusion. We also implemented an iterative feedback system that continuously aimed to refine and improve the annotation process. In cases of conflict or confusion, we removed those papers as we wanted only good quality dataset. Following the annotation phase, we obtained an average inter-annotator agreement score of 0.94 using Cohen’s kappa Cohen ( 1960 ) , indicating a substantial consensus among the annotators.

A.1.3 Annotator’s Pay

We compensated each annotator according to the standard PhD salaries in India, based on the hours they worked. The appointment and salaries adhere to our university’s established practices. Payment was made per paper since the time required to read and extract future research ideas from each paper varies, depending on its complexity, technical terminology, and the annotator’s familiarity with the subject. Thus, paying based on time spent could have potentially compromised the quality of the annotations. To maintain accuracy and prevent fatigue, we imposed a daily limit of six hours for annotators.

Appendix B Human Annotation

We prepared a Google Form for each paper and provided the links to the annotators. We also specified instructions for them at the beginning of the form. We have added an example of the form for a paper in Figure 10 , Figure 11 , and Figure 12 .

Here is the little modified from for human evaluation that generates only top 5 research ideas:-

Appendix C Effect of giving only Title and Abstract as Input

We found a few cases where we provided only an title and abstract as input to see if LLMs can still retain open-ended generation capabilities. We discovered few cases where GPT-4 still generated novel ideas, such as for a paper Kumar et al. ( 2023b ) it generated: "Incorporate explainable AI methods to provide transparency into how the AI model makes its predictions, thereby making the outcomes more interpretable and acceptable to human editors." . This kind of analysis has not been done yet and could be helpful. After providing full paper content to the model we found that same idea was again generated.

There were also cases where GPT-4 generated a novel idea of solving the problem using transformers for a task (The task was mostly solved using techniques like RNN), which had not been done before. However, after providing the full paper content, the model understood that this transformer has already been implemented for this task, so further suggested to add more contextual information to it to boost the result (limited information was given as input to the paper). Overall, we found that LLMs can still retain open-ended generation because it has past knowledge. But it may not generate many good ideas since it doesn’t have access to recently published papers or other methodological findings related to the current paper.

Appendix D Effect of Length of Idea on IAScore

In our analysis, we explore the relationship between the length of ideas and their corresponding Impact Assessment Score (IAScore), specifically focusing on computer science papers and outputs generated by GPT-4. This relationship is visually represented in the bar chart found in Appendix Figure 8 . The data reveal that shorter ideas, typically under 20 words, tend to receive lower IAScores. This could be attributed to their lack of detailed information, which might be essential for a comprehensive understanding and assessment. Conversely, we observe that ideas spanning 40-60 words also tend to score lower. This may result from their verbosity; excessive information can dilute the core message, making it challenging to discern the main points. Interestingly, ideas with a moderate length, ranging from 20 to 40 words, achieve the highest IAScores. This length seems optimal as it allows for sufficient detail without overwhelming the reader, striking a balance that facilitates clearer understanding.

Appendix E Effect of Adding Additional Background Knowledge

We designed our framework based on the Retrieval-Augmented Generation (RAG) model Lewis et al. ( 2020 ) to integrate background knowledge into LLMs, as illustrated in Figure 9 .

E.1 Vector Database

We utilized the Semantic Scholar API Kinney et al. ( 2023 ) to collect the titles and abstracts of approximately 1.9 lakh existing computer science research papers. We employed BERT embeddings to create vector representations for the titles of these papers, which were then stored in a vector database.

E.2 Retriever

To retrieve relevant papers, we created embeddings for the title of the paper for which we have to generate ideas. We computed the cosine similarity between this paper title embedding and those from our vector database. We then retrieved the top 20 research papers that exhibited the highest similarity to our target paper title. Finally, we extracted the contributions from these papers to gather relevant data from their abstracts.

We used the following prompt to instruct LLM to extract useful information from abstract of the paper:

We designed the above query prompt to ensure that the LLM 7 7 7 We employed Gemini-Pro model for this task understood its role in extracting relevant information without altering the provided information.

E.3 Generator

Next we produced the ideas using a prompt that includes the prompt using the paper and the retrieved background knowledge.

Specifically we used the below prompt for our task:-

Here, ‘[paper text]’ contains the full content of the paper after removal of future work sections. ‘[background knowledge]’ contains the background knowledge retrieved. An example of background knowledge is shown in Appendix Table 6 .

We performed this experiment on the same set of papers and conducted human evaluations for novelty following the same methodology as we discussed in Section 4.4 . The results are shown in Figure 7 . Initially, we observed that adding background knowledge affected the LLM’s performance; it primarily generated ideas that already existed, merely creating new combinations from the background knowledge. Subsequently, we modified the prompt to instruct the model not to repeat ideas that were mentioned in the background knowledge.

We found that adding background slightly improved the task. The results show that the improvements for GPT-4 and Claude were 50% and 53.33%, respectively, in reducing the generation of non-novel ideas. Also, it resulted in the improvement of 7.14% and 11.76% not generating generic ideas of GPT-4 and Claude. We observed that GPT-4 generated 9.52% and 14.63% more moderately novel ideas. However, we noted only a very slight improvement in the generation of highly novel or extremely novel ideas.

The analysis revealed that 73.71% of the ideas generated by Claude and 93.34% by GPT-4 were relevant. We observed that the relevancy score for Claude decreased by 2.96%, and GPT-4 increased by a slight 0.77%. Furthermore, 83.14% of Claude’s generated ideas and 96.98% of GPT-4’s ideas were judged to be practically feasible and factually correct. The score for Claude decreased by 0.20%, and the score for GPT-4 increased by 0.34%. It seems that additional information negatively impacts Claude’s performance by generating ideas that are irrelevant, non-novel, and infeasible. However, for GPT-4, we observed that incorporating additional background knowledge helps prevent the generation of non-novel or generic ideas and slightly improves the relevance and factual correctness of the generated ideas. However, further research is needed to enhance the ability of LLMs to generate more novel ideas.

Appendix F BERTScore Implementation Details

The motivation to use BERT embeddings is that the generated and the original ideas often do not use the same words, so we need to understand the contextual meanings of the ideas in order to compare them. We used the default setting of the BERTScore metric, which employs a 24-layer RoBERTa-large model and utilizes the 17th layer for embedding. We determined the threshold 8 8 8 We set the threshold 0.68 empirically using the validation set. If the similarity exceeds that threshold, we classify those pairs of ideas as similar, and vice versa.

Appendix G Error Analysis:

We conducted human evaluation using three expert annotators, each with over five years of experience in this field. They reviewed 15 papers. We assigned papers to each reviewer based on their familiarity with the subject matter of the papers. We identified two major reasons for the low IAS score:

Generic Ideas: Few ideas such as “Explore different explainability methods like LIME, SHAP to generate model explanations instead of just rationales. Compare their effectiveness." , Building on the baseline model, future research could explore more advanced natural language processing (NLP) models and techniques for contradiction detection. are generated. These statements are true; however, they are very generic and are common.

Author Miss: Due to page limits or more novel ideas, the author fails to mention a few ideas in a paper. For example, for a paper Kumar et al. ( 2023a ) GPT-4 generated idea: "Exploring the Impact of Contradictions on Review Outcomes: An interesting area for future research would be to study the impact of reviewer contradictions on the outcomes of the peer review process. This could involve analyzing the correlation between the presence and nature of contradictions and the final decisions made by editors (acceptance, rejection, major/minor revisions). Such studies could provide valuable insights into how contradictions influence the decision-making process and how they might be effectively managed to improve the fairness and quality of peer review." . This represents a strong, novel research problem not mentioned by the authors, which warrants future investigation

Appendix H Direct FRI and Mixed FRI

Direct FRI : When the sentences that mention future research idea only contains future research idea. For example "In future work, we plan to extend our approach to other code-mixed languages and evaluate its performance on more NLP tasks."

Mixed FRI : We found that sometimes research papers articulate future research ideas along with other essential information of the paper in a single sentence. For example in Figure 2 , this sentence not only summarizes the current research findings but also clearly outlines a direction for future work.

Appendix I Output Examples

Our LLM generated future research output can be found in Table 1 , Table 2 , Table 3 , Table 4 and Table 5 .

• Anil et al. (2023) Rohan Anil, Sebastian Borgeaud, Yonghui Wu, Jean-Baptiste Alayrac, Jiahui Yu, Radu Soricut, Johan Schalkwyk, Andrew M. Dai, Anja Hauth, Katie Millican, David Silver, Slav Petrov, Melvin Johnson, Ioannis Antonoglou, Julian Schrittwieser, Amelia Glaese, Jilin Chen, Emily Pitler, Timothy P. Lillicrap, Angeliki Lazaridou, Orhan Firat, James Molloy, Michael Isard, Paul Ronald Barham, Tom Hennigan, Benjamin Lee, Fabio Viola, Malcolm Reynolds, Yuanzhong Xu, Ryan Doherty, Eli Collins, Clemens Meyer, Eliza Rutherford, Erica Moreira, Kareem Ayoub, Megha Goel, George Tucker, Enrique Piqueras, Maxim Krikun, Iain Barr, Nikolay Savinov, Ivo Danihelka, Becca Roelofs, Anaïs White, Anders Andreassen, Tamara von Glehn, Lakshman Yagati, Mehran Kazemi, Lucas Gonzalez, Misha Khalman, Jakub Sygnowski, and et al. 2023. Gemini: A family of highly capable multimodal models . CoRR , abs/2312.11805.
• Anthropic (2023) Anthropic. 2023. Model card for claude 2 .
• Benedek et al. (2014) Mathias Benedek, Emanuel Jauk, Andreas Fink, Karl Koschutnig, Gernot Reishofer, Franz Ebner, and Aljoscha C. Neubauer. 2014. To create or to recall? neural mechanisms underlying the generation of creative new ideas . NeuroImage , 88:125–133.
• Brahman et al. (2023) Faeze Brahman, Chandra Bhagavatula, Valentina Pyatkin, Jena D Hwang, Xiang Lorraine Li, Hirona J Arai, Soumya Sanyal, Keisuke Sakaguchi, Xiang Ren, and Yejin Choi. 2023. : Making small language models better procedural knowledge models for (counterfactual) planning. arXiv preprint arXiv:2305.19472 .
• Bubeck et al. (2023) Sébastien Bubeck, Varun Chandrasekaran, Ronen Eldan, Johannes Gehrke, Eric Horvitz, Ece Kamar, Peter Lee, Yin Tat Lee, Yuanzhi Li, Scott M. Lundberg, Harsha Nori, Hamid Palangi, Marco Túlio Ribeiro, and Yi Zhang. 2023. Sparks of artificial general intelligence: Early experiments with GPT-4 . CoRR , abs/2303.12712.
• Carlini et al. (2022) Nicholas Carlini, Daphne Ippolito, Matthew Jagielski, Katherine Lee, Florian Tramèr, and Chiyuan Zhang. 2022. Quantifying memorization across neural language models . ArXiv , abs/2202.07646.
• Chen et al. (2021) Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Pondé de Oliveira Pinto, Jared Kaplan, Harrison Edwards, Yuri Burda, Nicholas Joseph, Greg Brockman, Alex Ray, Raul Puri, Gretchen Krueger, Michael Petrov, Heidy Khlaaf, Girish Sastry, Pamela Mishkin, Brooke Chan, Scott Gray, Nick Ryder, Mikhail Pavlov, Alethea Power, Lukasz Kaiser, Mohammad Bavarian, Clemens Winter, Philippe Tillet, Felipe Petroski Such, Dave Cummings, Matthias Plappert, Fotios Chantzis, Elizabeth Barnes, Ariel Herbert-Voss, William Hebgen Guss, Alex Nichol, Alex Paino, Nikolas Tezak, Jie Tang, Igor Babuschkin, Suchir Balaji, Shantanu Jain, William Saunders, Christopher Hesse, Andrew N. Carr, Jan Leike, Joshua Achiam, Vedant Misra, Evan Morikawa, Alec Radford, Matthew Knight, Miles Brundage, Mira Murati, Katie Mayer, Peter Welinder, Bob McGrew, Dario Amodei, Sam McCandlish, Ilya Sutskever, and Wojciech Zaremba. 2021. Evaluating large language models trained on code . CoRR , abs/2107.03374.
• Cohen (1960) Jacob Cohen. 1960. A coefficient of agreement for nominal scales. Educational and Psychological Measurement , 20:37 – 46.
• Devlin et al. (2019) Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: pre-training of deep bidirectional transformers for language understanding . In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2019, Minneapolis, MN, USA, June 2-7, 2019, Volume 1 (Long and Short Papers) , pages 4171–4186. Association for Computational Linguistics.
• Hao et al. (2020) Wenke Hao, Zhicheng Li, Yuchen Qian, Yuzhuo Wang, and Chengzhi Zhang. 2020. The ACL FWS-RC: A dataset for recognition and classification of sentence about future works . In JCDL ’20: Proceedings of the ACM/IEEE Joint Conference on Digital Libraries in 2020, Virtual Event, China, August 1-5, 2020 , pages 261–269. ACM.
• Huang et al. (2023) Qian Huang, Jian Vora, Percy Liang, and Jure Leskovec. 2023. Benchmarking large language models as ai research agents . ArXiv , abs/2310.03302.
• Kinney et al. (2023) Rodney Michael Kinney, Chloe Anastasiades, Russell Authur, Iz Beltagy, Jonathan Bragg, Alexandra Buraczynski, Isabel Cachola, Stefan Candra, Yoganand Chandrasekhar, Arman Cohan, Miles Crawford, Doug Downey, Jason Dunkelberger, Oren Etzioni, Rob Evans, Sergey Feldman, Joseph Gorney, David W. Graham, F.Q. Hu, Regan Huff, Daniel King, Sebastian Kohlmeier, Bailey Kuehl, Michael Langan, Daniel Lin, Haokun Liu, Kyle Lo, Jaron Lochner, Kelsey MacMillan, Tyler C. Murray, Christopher Newell, Smita R Rao, Shaurya Rohatgi, Paul Sayre, Zejiang Shen, Amanpreet Singh, Luca Soldaini, Shivashankar Subramanian, A. Tanaka, Alex D Wade, Linda M. Wagner, Lucy Lu Wang, Christopher Wilhelm, Caroline Wu, Jiangjiang Yang, Angele Zamarron, Madeleine van Zuylen, and Daniel S. Weld. 2023. The semantic scholar open data platform . ArXiv , abs/2301.10140.
• Kornish and Hutchison-Krupat (2017) Laura J. Kornish and Jeremy Hutchison-Krupat. 2017. Research on idea generation and selection: Implications for management of technology . Production and Operations Management , 26:633–651.
• Kuhn and Hawkins (1963) Thomas S. Kuhn and David Hawkins. 1963. The structure of scientific revolutions . American Journal of Physics , 31:554–555.
• Kumar (1996) R. Kumar. 1996. Research Methodology: A Step-by Step Guide for Beginners . Longman Australia.
• Kumar et al. (2023a) Sandeep Kumar, Tirthankar Ghosal, and Asif Ekbal. 2023a. When reviewers lock horn: Finding disagreement in scientific peer reviews. arXiv preprint arXiv:2310.18685 .
• Kumar et al. (2023b) Sandeep Kumar, Tirthankar Ghosal, and Asif Ekbal. 2023b. When reviewers lock horns: Finding disagreements in scientific peer reviews . In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing , pages 16693–16704, Singapore. Association for Computational Linguistics.
• Lewis et al. (2020) Patrick S. H. Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, and Douwe Kiela. 2020. Retrieval-augmented generation for knowledge-intensive NLP tasks . In Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual .
• Li et al. (2015) Jiwei Li, Michel Galley, Chris Brockett, Jianfeng Gao, and William B. Dolan. 2015. A diversity-promoting objective function for neural conversation models . ArXiv , abs/1510.03055.
• Liu et al. (2019) Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692 .
• Lo et al. (2020) Kyle Lo, Lucy Lu Wang, Mark Neumann, Rodney Kinney, and Daniel Weld. 2020. S2ORC: The semantic scholar open research corpus . In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics , pages 4969–4983, Online. Association for Computational Linguistics.
• Louviere et al. (2015) J.J. Louviere, T.N. Flynn, and A.A.J. Marley. 2015. Best-Worst Scaling: Theory, Methods and Applications . Cambridge books online. Cambridge University Press.
• Nadkarni et al. (2021) Rahul Nadkarni, David Wadden, Iz Beltagy, Noah A. Smith, Hannaneh Hajishirzi, and Tom Hope. 2021. Scientific language models for biomedical knowledge base completion: An empirical study . ArXiv , abs/2106.09700.
• OpenAI (2023) OpenAI. 2023. GPT-4 technical report . CoRR , abs/2303.08774.
• Plato et al. (2000) Plato, G.R.F. Ferrari, and T. Griffith. 2000. Plato: ’The Republic’ . Cambridge Texts in the History of Political Thought. Cambridge University Press.
• Qi et al. (2023) Biqing Qi, Kaiyan Zhang, Haoxiang Li, Kai Tian, Sihang Zeng, Zhang-Ren Chen, and Bowen Zhou. 2023. Large language models are zero shot hypothesis proposers . ArXiv , abs/2311.05965.
• Romera-Paredes et al. (2023) Bernardino Romera-Paredes, Mohammadamin Barekatain, Alexander Novikov, Matej Balog, M Pawan Kumar, Emilien Dupont, Francisco J. R. Ruiz, Jordan S. Ellenberg, Pengming Wang, Omar Fawzi, Pushmeet Kohli, Alhussein Fawzi, Josh Grochow, Andrea Lodi, Jean-Baptiste Mouret, Talia Ringer, and Tao Yu. 2023. Mathematical discoveries from program search with large language models . Nature , 625:468 – 475.
• Sarica et al. (2021) Serhad Sarica, Binyang Song, Jianxi Luo, and Kristin L. Wood. 2021. Idea generation with technology semantic network . Artificial Intelligence for Engineering Design, Analysis and Manufacturing , 35:265 – 283.
• Sarsa et al. (2022) Sami Sarsa, Paul Denny, Arto Hellas, and Juho Leinonen. 2022. Automatic generation of programming exercises and code explanations using large language models . In ICER 2022: ACM Conference on International Computing Education Research, Lugano and Virtual Event, Switzerland, August 7 - 11, 2022, Volume 1 , pages 27–43. ACM.
• Spangler et al. (2014) W. Scott Spangler, Angela D. Wilkins, Benjamin J. Bachman, Meena Nagarajan, Tajhal Dayaram, Peter J. Haas, Sam Regenbogen, Curtis R. Pickering, Austin Comer, Jeffrey N. Myers, Ioana Stanoi, Linda Kato, Ana Lelescu, Jacques J. Labrie, Neha Parikh, Andreas Martin Lisewski, Lawrence A. Donehower, Ying Chen, and Olivier Lichtarge. 2014. Automated hypothesis generation based on mining scientific literature . In The 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ’14, New York, NY, USA - August 24 - 27, 2014 , pages 1877–1886. ACM.
• Tshitoyan et al. (2019) Vahe Tshitoyan, John Dagdelen, Leigh Weston, Alexander Dunn, Ziqin Rong, Olga Kononova, Kristin A. Persson, Gerbrand Ceder, and Anubhav Jain. 2019. Unsupervised word embeddings capture latent knowledge from materials science literature . Nat. , 571(7763):95–98.
• Wang et al. (2023) Qingyun Wang, Doug Downey, Heng Ji, and Tom Hope. 2023. Scimon: Scientific inspiration machines optimized for novelty. arXiv preprint arXiv:2305.14259 .
• Wang et al. (2019) Qingyun Wang, Lifu Huang, Zhiying Jiang, Kevin Knight, Heng Ji, Mohit Bansal, and Yi Luan. 2019. PaperRobot: Incremental draft generation of scientific ideas . In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics , pages 1980–1991, Florence, Italy. Association for Computational Linguistics.
• Wei et al. (2022) Jason Wei, Maarten Bosma, Vincent Y. Zhao, Kelvin Guu, Adams Wei Yu, Brian Lester, Nan Du, Andrew M. Dai, and Quoc V. Le. 2022. Finetuned language models are zero-shot learners . In The Tenth International Conference on Learning Representations, ICLR 2022, Virtual Event, April 25-29, 2022 . OpenReview.net.
• Xu et al. (2023) Yi Xu, Shuqian Sheng, Bo Xue, Luoyi Fu, Xinbing Wang, and Chenghu Zhou. 2023. Exploring and verbalizing academic ideas by concept co-occurrence . In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , pages 13001–13027, Toronto, Canada. Association for Computational Linguistics.
• Yao et al. (2022) Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao. 2022. React: Synergizing reasoning and acting in language models. arXiv preprint arXiv:2210.03629 .
• Young (2019) J.W. Young. 2019. A Technique for Producing Ideas . Independently Published.
• Zhang* et al. (2020) Tianyi Zhang*, Varsha Kishore*, Felix Wu*, Kilian Q. Weinberger, and Yoav Artzi. 2020. Bertscore: Evaluating text generation with bert . In International Conference on Learning Representations .
• Zhao et al. (2023) Wayne Xin Zhao, Kun Zhou, Junyi Li, Tianyi Tang, Xiaolei Wang, Yupeng Hou, Yingqian Min, Beichen Zhang, Junjie Zhang, Zican Dong, Yifan Du, Chen Yang, Yushuo Chen, Z. Chen, Jinhao Jiang, Ruiyang Ren, Yifan Li, Xinyu Tang, Zikang Liu, Peiyu Liu, Jianyun Nie, and Ji rong Wen. 2023. A survey of large language models . ArXiv , abs/2303.18223.