---
name: User Research
description: Listen for problems, not feature requests.
---
# User Research

Users are geniuses at describing their problems and unreliable at prescribing solutions. "Build X" is a hypothesis to investigate, not a requirement to ship.

## Interview technique
- **Open questions.** "Tell me about the last time you…" — not "would you use feature X?" The latter invites fantasy; the former surfaces reality.
- **Probe the past, not the future.** Past behavior is evidence; future intent is imagination. "When did you last actually do this?" beats "would you do this?"
- **Listen for the workaround.** The spreadsheets, scripts, and manual steps people have built to cope reveal the real problem they're solving.
- **Follow the emotion.** Where they get frustrated, animated, or resigned — that's where the value is.

## Output
Patterns across many users, not anecdotes. One user's request is noise; the same pain in 8 of 10 interviews is signal.