Breaking design fixation with divergent concepts
You've anchored on the first layout that came to mind and can't see past it. NN/g describes AI as a "cybernetic teammate" for ideation and notes that teams augmenting ideation with AI outperformed those without. Asking for structurally different approaches — not just visual variations — is a cheap way to widen the option space before you commit to one.
You are a principal product designer running a divergent ideation session with me. Design problem: {{design_problem}}. Users and context of use: {{user_and_context}}. Hard constraints (platform, must-have functions, limits): {{constraints}}. Generate five structurally different concepts — different interaction models and information structures, not visual reskins of one idea. For each concept give: - The core idea in one sentence. - The interaction model (how the user moves through it), in words plus a simple ASCII wireframe. - Who it fits best and in what situation. - The main usability risk, named against a real principle (for example recognition over recall, error prevention, Hick's law). - The cheapest way to test whether it works. Include at least one deliberately unconventional option to stretch the range. Constraints: respect every hard constraint I listed. Do not claim any concept is "the best" — leave that judgment to me. Do not cite statistics, studies, or competitor features you cannot name and verify; if you invoke a pattern, name the design principle behind it rather than inventing evidence. Treat these as raw starting points, not finished designs.
Fill in your details and the prompt updates live — then copy.
**Concept 3 — Command-bar first (unconventional).** Core idea: skip the dashboard; users type what they want. [ search or type a command... ] recent: Re-run "Q2 churn" Export CSV Fits power users doing repeat tasks. Main risk: recognition over recall — new users don't know what to type; mitigate with visible suggestions. Cheapest test: a clickable prototype seeded with three commands, 5 users. **Concept 1 — Guided history list.** Lower ceiling, higher floor; best for occasional users. Risk: gets long fast — needs search and grouping...
The full workflow
- Frame the problem and list every real constraint before generating.
- Generate the five concepts and drop any that break a hard constraint.
- Sketch the two strongest yourself to make them concrete.
- Validate with real users — usability principles suggest, but only testing decides.
Watch out for
AI ideation regresses toward the average of its training data — common patterns, and often the accessibility debt baked into them. Use the output as raw material and validate with real users, not as a finished direction.
Do not feed unreleased product strategy or NDA-covered material into consumer AI tools to brainstorm; treat confidential designs the way you would confidential documents.
Where this comes from
Every use case on this site is grounded in real reports from working ux designers — not invented by us.