# Product Discovery

> Plan and synthesize customer interviews, map desirability, viability, and feasibility assumptions, and frame Jobs-to-be-Done. Use when understanding adoption, switching, churn, or prioritizing what remains unknown before defining a solution or validation test.

- Skill: `majesticlabs-dev/product-discovery` (Agent Skill, multi-file: 3 files)
- Install (CLI): `npx skillmds@latest add majesticlabs-dev/product-discovery`
- Raw SKILL.md: https://api.skillmd.com/api/skills/majesticlabs-dev/product-discovery/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Product & Planning
- Author: majesticlabs-dev (https://skillmd.com/u/majesticlabs-dev)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/majesticlabs-dev/product-discovery

---


# Product Discovery

## Boundary

Own discovery research, synthesis, and assumption framing. Stop at prioritized assumptions and evidence gaps; `product-validation` owns executable demand tests and the proceed, sharpen, or pause verdict. Do not treat interviews as market proof, lead participants, or jump from feature requests to requirements.

## Required Inputs

- Decision or hypothesis under investigation
- Target user, buyer, situation, and known current behavior
- Existing research, product data, constraints, and competing explanations
- Access to participants, consent requirements, privacy constraints, and research limitations

## Workflow

1. Write the decision and list current assumptions as desirability, viability, or feasibility.
2. Choose interviews, observation, existing-data analysis, or a research prototype based on the question and participant risk.
3. For interviews, choose a light interview plan or a traceable multi-interview round. Use [customer-interview-system.md](references/customer-interview-system.md) when goals, hypotheses, questions, participant records, and findings must remain linked across a research round.
4. Ask about recent concrete behavior, current workaround, trigger, cost, search, switching, and decision process.
5. Map the job: situation, functional progress, emotional and social dimensions, and push, pull, anxiety, and habit forces.
6. Capture verbatim evidence separately from interpretation and minimize personal data.
7. Synthesize repeated patterns, contradictions, segment differences, and negative cases.
8. Rank assumptions by impact if wrong, evidence weakness, and reversibility.
9. State what evidence is still required before solution definition or an executable validation test.

## Output

1. **Research question and assumption map**
2. **Interview or research plan**
3. **Evidence table and JTBD framing**
4. **Synthesis with contradictions and confidence**
5. **Traceability register and round status when using the multi-interview workflow**
6. **Prioritized assumptions and evidence gaps**

## Quality Gate

- Questions focus on behavior, not compliments or hypotheticals.
- Anecdotes are not called patterns.
- Feature requests are traced to underlying jobs.
- Interpretation rules are set before research begins.
- In a traceable round, material findings cite participant records and state sample limits.
- No demand test or product verdict is smuggled into discovery.

## Reference

Use [discovery-methods.md](references/discovery-methods.md) for lightweight interview and assumption worksheets. Use [customer-interview-system.md](references/customer-interview-system.md) for a traceable multi-interview round.