What Is Ideation in Design Thinking? Techniques and Process

Most teams think they know how to ideate. They book a meeting room, put sticky notes on a wall, and call whatever comes out of it a brainstorm. Then they pick the idea the most senior person in the room seemed keenest on and start designing. This is not ideation. It is consensus dressed up as creativity.

Ideation -- the structured creative phase of design thinking -- is where the gap between a mediocre solution and a genuinely good one usually opens up. Rush it or skip it and you will spend the rest of the project executing the wrong idea very efficiently.

Want to practise ideation on a real brief? UX Academy (myuxacademy.com)'s Beginner UX Design course runs live ideation sessions with working UX professionals. Cohort 1 starts 5 Sep 2026 -- reserve your place with a £99 deposit.

What ideation actually is

Ideation is the third stage of design thinking, sitting between Define (agreeing what problem to solve) and Prototype (making your best ideas tangible). Its job is to generate a wide range of possible solutions before any of them are evaluated.

The operative word is "before." Ideation works on a simple principle: diverge first, converge later. You open up the solution space as wide as possible, then narrow it down. Teams that collapse these two steps -- generating and evaluating simultaneously -- consistently produce narrower, less creative output than teams that keep them separate. This is well-established in the research on creative problem-solving -- the evidence consistently favours diverging before converging, generating options before evaluating them.

What ideation is not: a single brainstorming session, a vote on existing ideas, or a space where the loudest person wins. Done well, ideation is disciplined creative work with a clear structure and a deliberate output.

Where ideation fits in the UX process

If you are mapping this against the UX design process, ideation follows research and synthesis. You have spent time understanding your users, found patterns in what they told you, and arrived at a well-defined problem statement. Ideation is what you do next.

The sequence matters. Ideation without research produces solutions to problems nobody has. Research without ideation produces insight reports that change nothing. The two phases are co-dependent.

In the design thinking model specifically, the Define stage should hand ideation a sharp problem statement -- often framed as one or more "How Might We" questions. A good problem statement opens creative space without prescribing a solution. "How might we help new users understand the product's value before they commit to signing up?" is generative. "We need to redesign the onboarding flow" is not -- it has already decided the solution.

User personas built during research are an important input here. They stop ideation from becoming abstract. When a team is sketching ideas, asking "would Priya actually do this?" -- where Priya is a persona grounded in real interviews -- keeps solutions anchored to genuine human behaviour rather than designer assumptions.

Why ideation is not just brainstorming

Brainstorming is one tool. Ideation is the whole workshop.

The reason this distinction matters is that brainstorming has real weaknesses. In a group setting, people anchor to the first idea that gets said. Social pressure suppresses unusual or half-formed ideas. Extroverts dominate. The result tends to be a handful of ideas that cluster around whatever the group already thought before the session started.

Good ideation uses structures that get around these tendencies: silent individual generation before sharing, time pressure that forces quantity over polish, provocative constraints that break familiar patterns. The specific techniques matter less than the underlying principle: protect the generation phase from evaluation until there is enough volume to make comparison meaningful.

The other thing brainstorming misses is the value of individual divergence. Some of the most useful ideas in an ideation session come from one person working quietly for eight minutes, not from group energy. The techniques below mix both.

What are the best ideation techniques in design thinking?

How Might We

"How Might We" (HMW) is a question format that turns a problem statement into a creative prompt. "How might we reduce friction in the checkout flow?" invites solutions. "The checkout flow has too many steps" describes a problem without opening space for creative response.

HMW questions are often generated in clusters during synthesis: the team reviews their research themes and turns each one into a generative question. You might end up with ten to fifteen HMW questions from a single research synthesis session. Not all of them will be equally useful, but having a range of angles prevents the ideation from being dominated by the most obvious interpretation of the problem.

Crazy 8s

Crazy 8s is a time-boxed sketching exercise: eight rough sketches in eight minutes. The brief is to generate eight genuinely different ideas, not eight variations of the same idea.

The time pressure is deliberate. It removes the temptation to polish and forces volume. Most people produce ideas 6, 7, and 8 that they would never have reached without the constraint. It is also an equaliser -- in a room where some people are more comfortable speaking than sketching, Crazy 8s gives everyone the same format.

Sharing happens after the individual generation, with a short explanation of each sketch. The group discusses the ideas, not the quality of the drawings.

SCAMPER

SCAMPER is a structured checklist for generating variations on an existing idea or product. The letters stand for: Substitute, Combine, Adapt, Modify/Magnify, Put to other uses, Eliminate, Reverse/Rearrange. (borrowed from broader creative problem-solving practice, originally developed by Bob Eberle)

It is most useful when a team has a reasonable starting point but needs to push further. Working through each SCAMPER lens systematically generates modifications that the team would not have thought of in free-form discussion. "What if we eliminated this step entirely?" often surfaces a simplification that was staring the team in the face.

Worst Possible Idea

This technique inverts the brief: instead of generating good solutions, the team generates deliberately terrible ones. The value is in what those terrible ideas reveal about assumptions.

If the worst possible idea for improving a GP appointment booking system is "remove all confirmation messages," the inversion is obvious: confirmation is critical. But the group now has an explicit reason to explore what confirmation should look and feel like, rather than assuming the current approach is sufficient.

Worst Possible Idea also reduces the anxiety of early-stage idea generation. It is genuinely funny, which relaxes the room, and it shifts the group into a more generative mode before pivoting to serious solutions.

Design Studio

Design Studio is a structured session format for larger teams or complex problems. It typically runs in three rounds: individual silent sketching, pair or small-group critique, and whole-group convergence.

The structured critique element is what distinguishes it from straightforward sharing. Participants describe what they designed and why, then the group identifies what works and what questions remain -- without evaluating the person. Design Studio is time-intensive but produces more thoroughly stress-tested directions than shorter formats.

How to run an ideation session

Before the session: make sure the problem statement is agreed in advance. Sharing a HMW question or problem statement cold at the start of an ideation session wastes the first twenty minutes on discussion that should have happened during Define. Send it ahead. Brief participants on the format so they are not surprised by Crazy 8s.

Open with diverge: start with individual, silent generation -- either Crazy 8s or a quiet HMW response. This sets the norm that ideas are individual outputs first, not group productions. It prevents anchoring.

Share without judgement: go around the room and have each person present their sketches or ideas briefly. The facilitator's job is to record ideas visibly (on a shared whiteboard or Miro board) and prevent evaluation. "That won't work because..." is not allowed yet.

Cluster and build: once all ideas are visible, look for themes and combinations. Some of the best solutions in ideation sessions are hybrids of two ideas that seemed unrelated. Encourage building on ideas: "Yes, and..." rather than "but."

Converge: at the end, the group votes or discusses to identify two or three directions worth prototyping. The output should be a shortlist with enough shared understanding that someone can sketch a wireframe from it without needing to reconstruct the decision.

A session that ends with one "chosen" idea has usually converged too early. Two or three directions going into prototyping is the right output -- prototyping quickly will tell you more than any amount of in-room debate.

Common ideation mistakes

Evaluating too early. The most common problem. Someone says "that's interesting but we'd never get sign-off on it" and the idea dies. All evaluation, including feasibility filtering, should happen after generation is complete.

Only doing it once. Ideation is not a single event in a project lifecycle. It happens at multiple points: early to generate solution directions, again when prototypes reveal new constraints, again after usability testing surfaces problems the original solutions did not address. Teams that treat ideation as a one-off phase tend to over-invest in their first ideas and under-respond to what they learn later. Structured formats like design sprints build in multiple ideation phases by design, which is part of why they tend to produce stronger solutions than unstructured project workflows.

Ignoring the research. Ideation that is not grounded in user research produces solutions to problems the team imagined rather than problems users actually have. The personas, themes, and problem statements from synthesis are not optional inputs -- they are the reason the solutions generated in ideation will be relevant to anyone outside the room.

Using only one technique. Different people generate ideas in different ways. Some need visual prompts. Some need time pressure. Some need the structure of SCAMPER. Relying on a single format advantages whoever is naturally suited to it and disadvantages everyone else.

Skipping silent generation. Group discussion from the start creates anchoring. The first idea said out loud shapes everything that follows. Silent individual generation before sharing is not a nice-to-have; it is the structural element that makes the rest of the session work.

Ideation as a learnable skill

Ideation feels uncomfortable at first. Most people have spent years in educational and professional settings where half-formed ideas are implicitly penalised. The discipline of generating before evaluating runs counter to that instinct.

The good news is that it is genuinely learnable. Designers who run regular ideation sessions -- and who reflect on what worked and what did not -- get better at it. They learn which techniques unlock the room, how to facilitate without dominating, and how to protect the generation phase when someone tries to skip it.

That is one of the reasons live, supervised practice matters more than self-study in UX education. Reading about Crazy 8s gives you the concept. Actually running one in a structured session -- with feedback from an experienced facilitator who has run it in a professional setting -- gives you the skill.

If you are switching careers to UX design and wondering whether this kind of thinking will come naturally to you: it does not need to. It needs to be practised. That is what design education is for.

On UX Academy's Beginner UX Design course, ideation sessions are part of the live curriculum -- not a section to read through, but a structured workshop you participate in, with a brief grounded in real user research and lead instructor Natalia Veretenyk and her team, who have run these sessions professionally at Adobe, Google, and Canva. If you want to see what live UX training feels like before committing, the free UX masterclass at UX Academy (myuxacademy.com) is the place to start. If you are serious about doing this work professionally, learning to ideate well is one of the most transferable skills you can build. See the course details and reserve your place here.