UX Design Ideation

Ideation is where I move from understanding to options. I aim to generate lots of directions or ideas first, then converge quickly so we can test meaningful prototypes with users.

I keep sessions timeboxed, and treat every idea as a hypothesis we can learn from.

How Might We (HMW)

• I start from the problem statement (user + need + insight), then convert it into multiple “How Might We” questions that invite action.
• I make the prompts specific enough to guide ideation, but open enough to produce variety, for example: “How might we help [user] achieve [outcome] despite [barrier]?”
• If ideas cluster too early, I reframe: change the perspective (user, carer, stakeholder), swap the constraint, or ask “what would a beginner/ power user do?”.
• For each HMW, I aim for 5–8 distinct idea directions before we start selecting.

Crazy-8s (8 ideas in 8 minutes)

• I pick one priority HMW and timebox the session (usually 8 minutes) to stop overthinking.
• I sketch 8 panels quickly, focusing on different approaches to the same underlying need (not incremental UI tweaks).
• I allow “wild” options on purpose (new journeys, alternative information structures, different interaction styles) so we don’t default to the first safe idea.
• After the timer, I circle the 1–3 panels that feel most aligned with user value, inclusion, and testability, then translate them into short concept statements.

Filtering and refining

• Dot-voting (or lightweight impact/effort checks) to choose the directions we’ll prototype next.
• Affinity mapping to group related ideas into themes, then naming the emerging direction.
• Quick sanity checks: does it solve the user problem, is it understandable, and can we learn something by prototyping it?

Other ideation techniques I use

• Brainwriting (for example the 6-3-5 method) to avoid groupthink and encourage quieter participants.
• SCAMPER to generate variations: substitute, combine, adapt, modify, put to another use, eliminate, reverse.
• Mind mapping / topic webs from key insights to surface relationships between needs, emotions, and tasks.
• Analogies: borrow patterns from other domains, then map features back to the user context.
• Role storming / perspective taking to stress-test assumptions from different stakeholder or user viewpoints.

Select a technique to open details and resources.

Typical outputs

• 2–4 direction statements ready for prototyping.
• Associated “what we expect to learn” notes so testing targets the riskiest assumptions.

Goal: move from possibilities to testable concepts without getting stuck debating preferences.