UX Research Interview Kit

Say this to activate the skill

---
name: ux-research-interview-kit
description: Builds guides, screeners, and analysis frameworks for structured qualitative UX interviews.
---

This skill produces complete kits for qualitative UX interviews: screener, interview guide, note-taking template, and analysis grid.

## When this skill activates
- Customer interviews are coming up (relaunch, product decision, discovery) and a structured guide is needed.
- Interviews are done and notes or transcripts need to be coded and synthesized by theme.
- A question list should be audited for leading questions and methodological flaws.

## Workflow
1. Clarify the research question and the decision it informs. Any question that does not feed that decision gets cut.
2. Define the target segment and write a screener of 4 to 6 questions: intended participants in, pure opinion givers out. Rule of thumb: 5 to 8 interviews per segment.
3. Structure the guide as a funnel: warm-up (2 personal questions), context (current behavior), core (specific past episodes), deep dives (follow-up bank), closing.
4. Rewrite every question into behavioral form: "Tell me about the last time you did X" instead of "Would you use X?". Cut hypotheticals.
5. Run a leading-question check: test whether each question suggests a desired answer; remove loaded adjectives and hidden assumptions.
6. Test the time budget: 45 to 60 minutes, at most 10 to 12 main questions, one page, follow-ups marked as optional prompts.
7. Prepare the analysis grid up front: derive starter codes from the research question; capture observation, quote, and interpretation separately.
8. Synthesize: count a pattern as an insight only after 3 independent mentions, flag single voices, document contradictions.

## Output format
One Markdown document with four sections: 1) screener with selection logic, 2) interview guide (one page, time per block, follow-up bank), 3) note-taking template (timestamp, observation, quote, interpretation), 4) analysis grid with starter codes and an insight template (insight, supporting quotes, segment, implication).

## Quality rules
- No core question can be answered with yes or no; closed questions belong only in the screener.
- Questions target past behavior, never opinions or hypothetical usage.
- No feature questions in discovery interviews; solution feedback belongs in usability tests.
- Every insight is backed by at least one verbatim quote.
- Observation and interpretation stay strictly separated.
- The guide still works when the question order changes.
- The report states participant count and segment.

Share this skill