Brainstorming

Brainstorming

Turn vague ideas into concrete, well-formed plans through structured collaborative dialogue. This skill applies to any creative PM work: feature ideation, experiment design, strategy formation, or solution exploration.

When to Use

• "brainstorm ideas for..."
• "help me think through..."
• "I have an idea but it's not fully formed"
• "explore options for..."
• "what should we build for..."
• Before starting a PRD, experiment, or new initiative

Process

Phase 1: Understand the Problem

Ask one question at a time to understand:

1. What's the context? — Product area, current state, what prompted this
2. Who's the customer? — Reference knowledge/_template/customer-segments.md to ground in real personas
3. What's the problem? — Frame as: I am... I am trying to... But... Because... Which makes me feel...
4. What does success look like? — Metric, outcome, or state change
5. What are the constraints? — Timeline, resources, technical limitations, regulatory

Phase 2: Explore Approaches

1. Generate 3 options ranging from conservative to ambitious
2. For each option, assess:

• Customer impact — How much does this move the needle for the persona?
• Effort — Engineering, design, ops complexity
• Risk — What could go wrong? Reversible or irreversible?
• Strategic fit — Does it ladder to a strategic pillar? (Reference knowledge/_template/strategy.md)

1. Present a recommendation with reasoning

Phase 3: Sharpen the Idea

1. Pressure-test the recommended approach:

• What's the hypothesis? ("We believe [action] will [outcome] for [segment] because [rationale]")
• What's the simplest version we could ship to learn?
• What would make us kill this idea?

1. Define next steps: PRD, experiment, one-pager, or prototype

Output

A brainstorm summary:

• Problem (customer problem framework)
• Options explored (3, with trade-offs)
• Recommendation (with rationale)
• Hypothesis (testable)
• Next step (what to do now — and which skill to use: prd-writer, experiment-designer, one-pager)

Principles

• One question at a time — don't overwhelm
• Multiple choice when possible — easier to react to options than generate from scratch
• YAGNI — cut unnecessary scope ruthlessly
• Bias toward experiments — if unsure, design an experiment, don't debate
• Customer first — always ground in a real persona and real problem

Knowledge References

• knowledge/_template/customer-segments.md — customer personas and segments
• knowledge/_template/strategy.md — strategic alignment
• knowledge/_template/metrics-and-targets.md — success metrics

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: hardiktiwari
• Source: hardiktiwari/PM-operating-OS
• License: MIT

Install and usage instructions live in the source repository linked above.