Rapid Ideation separates divergent generation from evaluation. A clear how-might-we question, visible constraints, and silent individual work reduce group conformity.

1. 1

   Focus

   Focus

   

   Goal

   Produce ideation brief so the core question “Which distinct solution directions address the focused user barrier?” can be answered with stronger support.

   Inputs for this step

   • How-might-we question
   • research evidence
   • constraints and no-gos

   Actions

   • Expose HMW question, user barrier, and no-gos.
   • Record source, rationale, and open uncertainty directly in ideation brief.

   Output

   ideation brief

   ideation brief · source · uncertainty · next decision

   Quality check

   Is exactly one barrier in focus?

   Pitfall

   A broad question yields variants of one solution and early voting removes unusual options. Sharpen question and constraint, diverge silently again, and apply criteria only after clustering.

   Timing: 10–20 minutes

2. 2

   Diverge

   Diverge

   

   Goal

   Produce raw ideas so the core question “Which distinct solution directions address the focused user barrier?” can be answered with stronger support.

   Inputs for this step

   • ideation brief
   • research evidence
   • constraints and no-gos

   Actions

   • Write or sketch ideas in a short silent timebox.
   • Record source, rationale, and open uncertainty directly in raw ideas.

   Output

   raw ideas

   raw ideas · source · uncertainty · next decision

   Quality check

   Has each person generated several directions?

   Pitfall

   A broad question yields variants of one solution and early voting removes unusual options. Sharpen question and constraint, diverge silently again, and apply criteria only after clustering.

   Timing: 10–20 minutes

3. 3

   Cluster

   Cluster

   

   Goal

   Produce idea landscape so the core question “Which distinct solution directions address the focused user barrier?” can be answered with stronger support.

   Inputs for this step

   • raw ideas
   • research evidence
   • constraints and no-gos

   Actions

   • Cluster by mechanism and mark unusual options.
   • Record source, rationale, and open uncertainty directly in idea landscape.

   Output

   idea landscape

   idea landscape · source · uncertainty · next decision

   Quality check

   Are clusters based on mechanism rather than author?

   Pitfall

   A broad question yields variants of one solution and early voting removes unusual options. Sharpen question and constraint, diverge silently again, and apply criteria only after clustering.

   Timing: 10–20 minutes

4. 4

   Select

   Select

   

   Goal

   Produce test candidates so the core question “Which distinct solution directions address the focused user barrier?” can be answered with stronger support.

   Inputs for this step

   • idea landscape
   • research evidence
   • constraints and no-gos

   Actions

   • Select contrasting options by problem fit and learning value.
   • Record source, rationale, and open uncertainty directly in test candidates.

   Output

   test candidates

   test candidates · source · uncertainty · next decision

   Quality check

   Will candidates produce different learning?

   Pitfall

   A broad question yields variants of one solution and early voting removes unusual options. Sharpen question and constraint, diverge silently again, and apply criteria only after clustering.

   Timing: 10–20 minutes