| name | user-research |
| description | Use when the user wants to understand their customers, design user interviews, synthesize research findings, or validate assumptions about user behavior. Also use when the user mentions 'user research,' 'customer interviews,' 'qualitative research,' 'user pain points,' 'personas,' 'customer discovery,' or 'understanding users.' |
User Research
User research is not asking users what they want. It is uncovering what they struggle with, why they behave as they do, and what outcomes they are actually trying to achieve.
Research Types
Generative research — Discover unknown problems. Use early, before you have a solution.
Evaluative research — Test a hypothesis or design. Use when you have something to test.
Most teams skip generative research and wonder why their features don't land.
Interview Guide Structure
Opening (5 min)
- "Tell me about your role and what a typical day looks like."
- "Walk me through the last time you did [task relevant to your product]."
Problem Exploration (20-30 min)
- "What's the hardest part of [task]?"
- "What do you do when that happens?"
- "How do you handle it today?"
- "What does it cost you when that goes wrong?" (time, money, frustration)
Closing (5 min)
- "Is there anything I should have asked that I didn't?"
- "Who else would you suggest I talk to?"
Interview Rules
- Never ask "would you use X?" — behavior, not hypothetical preference
- Never lead with the solution — problem first, always
- Follow up with "tell me more" and "why?" before moving on
- Record with permission — notes miss nuance
Synthesis: Finding Patterns
After 5+ interviews:
- Dump — One note per observation, quote, or behavior
- Group — Cluster by theme (no labels yet)
- Name — Label each cluster with an insight statement
- Rank — Which clusters appear in 3+ interviews? Those are patterns.
Insight statement format:
"[Segment] struggles with [problem] when [context] because [root cause], which means [cost/outcome]."
Sample Sizes
| Purpose | Minimum interviews |
|---|
| Generative discovery | 5-8 (patterns emerge fast) |
| Validating a hypothesis | 5 (stop if pattern is clear) |
| Persona construction | 15-20 per segment |
| Usability testing | 5 (reveals ~85% of usability issues) |
Common Rationalizations
| Rationalization | Reality |
|---|
| "We already know our users" | You know who they are. Research tells you why they do what they do. |
| "We'll do research after we launch" | Post-launch research tells you what broke. Pre-launch research prevents it. |
| "5 interviews isn't statistically significant" | Qualitative research isn't statistics. It's pattern detection. 5 is enough. |
| "Our NPS data tells us what users think" | NPS tells you satisfaction. Not why, and not what to do about it. |
Verification
