Market Problems — discovery & validation (Framework: Market → Market Problems)

Rule: The answers to most of your questions are not in the building. Discover, collect, analyze and communicate problems found in the market, then validate they are urgent, pervasive, and that the market is willing to pay to solve them. See ../pm-copilot/references/framework.md for the full map.

Who to listen to

The Noisy 20% vs. the Quiet 80%. The noisy 20% (your most excited/unhappy customers plus loud internal voices — an exec's advisory board, a big-client sales rep, support) give a limited, skewed view. The quiet 80% won't come to you; you must go find them (industry conferences — leave your booth; professional associations; LinkedIn; end-user training). Noisy customers don't represent the whole market.

The four market segments — each is "a person with a problem." Balance interviews across all four:

NIHITO — Nothing Interesting Happens In The Office. Recommend ~10 market visits per product manager per quarter. It feels like cold calling, but you aren't selling — state what's in it for them and expect many "no"s.

Discovery techniques

Observation — take thorough notes, look for patterns, minimize interruptions; if something catches your eye, ask. Answer: What do customers do / how / why? What's done manually that could be automated? What's multi-step that could be streamlined? How do they interact with your product? When do they use other systems, search for info, or need help?

Interview — prepare beforehand; use high-gain, open-ended questions; spend most of the time listening, not educating. Ask: What obstacles keep you from your goals? How are your goals measured? What do you think about our product/company? What else have you considered? How bad is the problem? How bad do things have to get before they're fixed? Where does this fall on your priorities?

Validate the problem — the three-part test

An opportunity is the intersection of all three (fail any one → don't build):

• Urgency — When is it urgent? Where is the urgency coming from? What's the level?
• Pervasiveness — Is it big enough to build a business on? (look for the quiet 80%)
• Willingness to pay — Is it painful enough to pay to solve? Who will pay — and will they pay you?

Validate quantitatively with surveys (close-ended questions on the three criteria — avoid leading questions) and experimentation (inexpensive, low-fidelity tests of your hypothesis).

Interview the user (batch questions)

1. What product/feature area, and who do you think has the problem (persona/segment)?
2. What evidence do you already have (interviews, tickets, win/loss, usage, surveys)?
3. Which of the four segments have you actually talked to? Which are you missing?
4. For each candidate problem: how urgent, how pervasive, would they pay — and what's the proof?

Artifact templates

Market Discovery Document (one per visit/interview — one file each, named

01-market-discovery-<contact-or-org>.md per artifact-output.md)

# Market Discovery Document

**Company:** <company>  ·  **Feature / Product:** <feature / product name>
**Author:** <author>  ·  **Date created:** <date>  ·  **Version:** 1.0

## Contact Info
- **Name:**  |  **Title:**  |  **Organization:**
- **Work phone:**  |  **Email:**  |  **Website:**

## Product Info
- **Product name:**

## Purpose of Call
<why you're talking to them>

## Call Notes
<what they said — verbatim where possible>

## Key Observations and Problems
<problems in the persona's first-person voice; note urgency / pervasiveness / willingness-to-pay signals; which segment>

Market Problems Table (roll-up across discovery)

# Market Problems Table — <product / market>

**Company:** <company>  ·  **Feature / Product:** <feature / product name>
**Author:** <author>  ·  **Date created:** <date>

| Persona | Problem Name (first person) | Market Evidence | Impact (1–5, 99999=contractual) | Priority (Evidence × Impact) | Group |
| :--- | :--- | ---: | ---: | ---: | :--- |
| <persona> | "I can't …" | <count of inputs> | <1–5> | <=E×I> | <theme> |

**Market Evidence** = tally of documented occurrences across inputs (interviews, surveys,
sales calls, support tickets, win/loss, analyst reports, reviews, usage data).
Sort by Priority descending. Problems that clear urgent + pervasive + willing-to-pay
become candidate requirements.

Deliver the artifact

Follow ../pm-copilot/references/artifact-output.md: confirm inputs (flag thin evidence and recommend specific NIHITO visits to fill it), ask Markdown or .docx, write the .md, and convert to .docx on request via your environment's native document-creation capability. Then offer the next stage: define pm-personas for the people who have these problems, or write pm-positioning around the validated problems.