Ideation
Generate enough distinct options that the first idea has to compete, then converge deliberately rather than by exhaustion.
Definition
Ideation is the divergent step: producing several genuinely different approaches to a framed problem before committing. Its purpose is not creativity for its own sake, it is avoiding the first plausible answer becoming the only one considered.
Mixed evidence. Real research exists, but findings conflict, effect sizes are small, or the jump from the lab to interface design is contested.
On this plate
Reach for this when
- A problem is framed and more than one approach is plausible.
- The team has converged suspiciously quickly.
- A redesign is being justified by 'it has always worked this way'.
- A team has been iterating on a single idea for long enough that alternatives have stopped being imagined.
Common mistakes
- Ideating before the problem is framed.
- Group generation with no independent phase.
- Selecting by seniority or by volume of enthusiasm.
- Judging during generation, which quietly stops anyone offering the odd idea that the good one usually comes from.
Formats that produce different options
Group brainstorming is the weakest of these. Independent generation before discussion consistently produces more and more varied ideas.
- Individual sketching against a shared frame, then a silent review before anyone speaks.
- Time-boxed variations: eight ideas in eight minutes, which prevents polishing the first one.
- Deliberate constraints: solve it with no new screens, or with one screen, or without the word search.
- Precedent review: how five other products solved the same problem, used to widen rather than to copy.
- Inversion: design the worst possible version, then read the list as a checklist of what to avoid.
Running it
Do
- Share the frame and the constraints before anyone generates anything.
- Generate alone, review together. Anchoring in a group kills variation fast.
- Keep the weird options in the set until the comparison, so they can influence the safe ones.
- Converge against the stated success measure, not against whose idea it was.
Do not
- Do not critique during generation.
- Do not let the most senior person go first.
- Do not converge by fatigue. Name the criteria before choosing.
- Do not skip this step because the answer seems obvious. That is exactly when the first idea goes unchallenged.
A session that produces real options
- 1
Send the problem in advance
People think better with a day than with eight minutes, and the quiet ones arrive with something written.
- 2
Generate alone first
Individual sketching before any discussion. Groups that start by talking converge on the first idea spoken.
- 3
Show without attribution
Pin every sketch up. Who drew what changes how it is judged, and nothing else about it.
- 4
Force variety before quality
Ask for the cheap version, the ambitious version and the version that removes the feature entirely.
- 5
Close with a decision about what to test
A session that ends in a wall of ideas and no next step is where ideation earns its reputation.
In practice
- i.
The option nobody had considered
Why the removal version mattersAsking for a version that deletes the feature rather than improving it produced the shipped answer: the setting nobody could design well turned out to be one nobody needed to make.
Local chart
FLOW·008 · Region IV · Grid G2
This entry and the entries it is stated to connect with. Each line is labelled with what the connection means.
Sources
Where a source establishes something narrower than the popular reading of it, the note says so.
- [1]
Productivity loss in brainstorming groups
Diehl and Stroebe, 1987
Experimental evidence that nominal groups generate more ideas than interacting groups, the basis for generating alone first.
Continue from here
Each link says what the connection is, so you can tell a principle from an alternative from a thing people mix this up with.
Principles behind this
The reasoning this solution is an application of.
- Anchoring BiasBIAS·001 · Cognitive bias
Related concept
Connected closely enough to change how you apply this.
- Design CritiqueFLOW·001 · Workflow
- Problem FramingFLOW·007 · Workflow
- Prototyping FidelityFLOW·005 · Workflow
- User FlowsFLOW·009 · Workflow
- WireframingFLOW·010 · Workflow