The brainstorm list is the simplest and most misused ideation format. Done right, it's a quantity game: defer judgment, keep the ideas coming, and evaluate only after the board is full. Snap it and let BoardSnap sort the signal from the noise.
Run a brainstorm at the start of any ideation phase — before a design sprint, before a planning session, before a product strategy discussion. The brainstorm is generative, not evaluative. Use it to build the raw material that other methods (crazy eights, ideation grid, impact/effort matrix) will refine.
Brainstorming only works when the evaluation phase is strictly separated from the generation phase. The moment someone says 'that won't work,' the brainstorm is over.
The question the brainstorm is answering. Write it prominently at the top of the board before any ideas are generated. A good challenge statement is specific enough to constrain the ideas but open enough to invite diverse approaches. 'How might we reduce the time from whiteboard to action item to under 30 seconds?' produces better ideas than 'How might we improve our product?'
The first pass — everyone generates ideas independently and posts them on the board. One idea per sticky. No talking about ideas yet. No evaluation. Aim for at least 20 ideas in round 1. Quantity is the goal; originality will come in later rounds.
After reading all round 1 ideas, a second pass where participants add ideas inspired by what they see. Round 2 ideas often build on, combine, or invert round 1 ideas. This is where the most creative ideas typically appear — they're built from the mental landscape of what's already on the board.
A designated area for ideas that are explicitly impossible, impractical, or absurd. Wild ideas aren't meant to be implemented directly — they're meant to trigger adjacent practical ideas. 'What if the whiteboard transcribed itself?' might lead to 'What if we added a voice-to-sticky feature?' Give wild ideas their own space so they don't get suppressed.
Only after the board is full: group related ideas into clusters. Evaluate clusters, not individual ideas. The question is 'which direction is most worth exploring?' not 'which specific idea is best?' Dot vote on the top clusters and advance those to the next stage.
Write the challenge statement. State the two rules: one idea per sticky, no evaluation during the generation phase. Assign a timekeeper. The rules must be enforced — the first person who says 'that won't work' ends the brainstorm.
Everyone writes individually and posts stickies on the board. Silence prevents anchoring — the first person to speak shapes what everyone else thinks about. Silent individual writing produces more diverse ideas than open group discussion.
Someone reads every sticky aloud. No commentary. Just reading. This gives everyone a complete picture of what's been generated before round 2 begins.
Second round of silent writing. Build on, combine, or invert ideas from round 1. Post the wild ideas in the wild ideas zone. Aim for 50% more stickies than round 1 produced.
Group related stickies into clusters. Name each cluster. Dot vote on the top 3 clusters. These become the input for the next session — crazy eights, ideation grid, or solution sketching.
Snap the full board with BoardSnap before the stickies are removed. The AI reads every idea and outputs a structured cluster summary — preserving the ideas that didn't make it into the top clusters, which might be valuable in a future session.
A brainstorm board is the most ephemeral team artifact there is — stickies fall off, the board gets erased, and 90% of the generated ideas disappear without a trace. Most of those ideas deserved to survive longer than they did. An idea that didn't make it into the top clusters today might be exactly what's needed three sprints from now.
BoardSnap captures every sticky — the winners and the also-rans — in a structured inventory. The AI clusters the ideas by theme and preserves the full idea set as a searchable artifact. Ideas don't die when the board is erased; they live in BoardSnap until a future session needs them.
The research on this is nuanced. Nominal group technique (individuals brainstorm independently, then share) consistently outperforms interactive group brainstorming in idea quantity and quality. The version described here uses silent individual writing followed by sharing — which captures the benefits of both: individual generation without anchoring, plus social building in round 2. The critical rule is strict separation of generation from evaluation.
30–45 minutes for a typical session with 4–8 participants. Shorter and you don't get past the obvious ideas. Longer and idea quality starts to decline. Two round-robin passes of 8 minutes each, separated by a read-all phase, is the practical optimum for most team sizes.
Evaluation during generation. 'That won't work,' 'we tried that,' 'that's too expensive' — any evaluative statement during the generation phase suppresses subsequent ideas. The facilitator must interrupt evaluative comments immediately and redirect to generation mode. Build a separate evaluation phase into the session structure so participants don't feel the need to evaluate during generation.
Before. Giving participants 24–48 hours to think about the challenge before the session consistently produces more and better ideas. The best ideas in a brainstorm are rarely the ones generated under pressure in the room — they're the ones that came during the morning commute after the participant read the challenge the night before.
No exporting, no transcription. Snap the board, get the action plan.