AI Brainstorm Sprint
Generate 50 ideas in 20 minutes, then find the three worth pursuing.
A brainstorm sprint uses AI to generate a large, uncensored set of ideas fast, then applies structured filtering to find the ones with genuine potential. The AI is for quantity; you're for quality.
The Workflow
Define the challenge clearly
A good brainstorm prompt is a well-formed problem statement. Vague challenges produce generic ideas. Specific constraints actually help creativity.
Run the divergent brainstorm
Generate a large volume of ideas without filtering. Quantity first, quality later.
Run a brainstorm sprint on this challenge: [CHALLENGE_STATEMENT] Context: - Audience: [TARGET_AUDIENCE] - Constraint (what makes this hard): [MAIN_CONSTRAINT] - What success looks like: [SUCCESS_DEFINITION] - What I've already tried or ruled out: [ALREADY_TRIED] Generate 40 ideas. Don't filter for quality — I want the obvious ideas, the weird ideas, and the ideas that might be impossible. Group them loosely into: - Conventional approaches (10) - Variations and twists on those (10) - Contrarian approaches (10) - Wild ideas (10) Don't explain or defend any of them yet. Just list them.
Replace: [CHALLENGE_STATEMENT], [TARGET_AUDIENCE], [MAIN_CONSTRAINT], [SUCCESS_DEFINITION], [ALREADY_TRIED]
Filter and develop the best ideas
Now apply judgment. Identify the most promising ideas and develop them further.
From the brainstorm we just ran, I've starred these ideas as most interesting to me: [SELECTED_IDEAS] For each starred idea: 1. Identify why it might work — the core insight behind it 2. Identify the biggest obstacle to making it real 3. Give me the simplest possible version I could test in one week with minimal resources 4. Name one person, company, or project that has done something similar I could learn from Then tell me: which of these ideas do you think has the highest chance of actually working given [MAIN_CONSTRAINT]? Be honest.
Replace: [SELECTED_IDEAS], [MAIN_CONSTRAINT]
All Prompts for This Workflow
Brainstorm sprint on this challenge: [CHALLENGE_STATEMENT] Audience: [TARGET_AUDIENCE] Constraint: [MAIN_CONSTRAINT] Success looks like: [SUCCESS_DEFINITION] Already tried: [ALREADY_TRIED] Generate 40 ideas without filtering. Group into: conventional (10), twists (10), contrarian (10), wild (10). No explanations yet — just list them.
Replace: [CHALLENGE_STATEMENT], [TARGET_AUDIENCE], [MAIN_CONSTRAINT], [SUCCESS_DEFINITION], [ALREADY_TRIED]
I have these two separate ideas I like but haven't been able to merge: Idea A: [IDEA_A] Idea B: [IDEA_B] Generate 10 combinations or hybrids that take elements from both. For each combination: - Describe it in one sentence - Name the core tension it introduces - Rate its originality 1-5 Then pick the two most promising hybrids and sketch what a first version would look like.
Replace: [IDEA_A], [IDEA_B]
I need to [CREATIVE_GOAL] but I have these constraints: - [CONSTRAINT_1] - [CONSTRAINT_2] - [CONSTRAINT_3] Instead of working around these constraints, use them as creative parameters. Generate 15 ideas that treat each constraint as a feature, not a bug. Then: what constraint, if removed, would produce the most boring version of this? That tells us which constraint is most creatively productive to keep.
Replace: [CREATIVE_GOAL], [CONSTRAINT_1], [CONSTRAINT_2], [CONSTRAINT_3]
40+ ideas organized by approach type, followed by developed versions of your selected ideas including a simplified test version, comparable examples, and an honest assessment of the most viable option.
- →Don't filter during the divergent phase. Bad ideas are often what eventually become good ideas when twisted slightly.
- →After filtering, ask AI: 'Which idea on my rejected list would work best if I had unlimited budget?' — sometimes the version of an idea is the obstacle, not the idea.