
user-research
by nicepkg
🚀 170+ pre-built skills for Claude Code, Cursor, Codex & 14+ AI tools. Stop re-teaching your AI the same things. One command → instant domain expertise. Marketing, SEO, Trading, Video, PM workflows included.
SKILL.md
name: user-research version: "2.0.0" description: User research methods, customer insight gathering, and problem validation for product discovery. sasmp_version: "1.3.0" bonded_agent: 02-discovery-research bond_type: PRIMARY_BOND parameters:
- name: research_method type: string enum: [interviews, surveys, usability, observation, analytics] required: true
- name: sample_size type: number default: 20 retry_logic: max_attempts: 3 backoff: exponential logging: level: info hooks: [start, complete, error]
User Research Skill
Conduct effective user research to understand customer needs, behaviors, and pain points. Master interview techniques and insight synthesis.
Research Methods
Method Selection Guide
| Method | When | Sample | Duration |
|---|---|---|---|
| Interviews | Deep understanding | 15-25 | 45-60 min |
| Surveys | Quantitative validation | 100+ | 5-10 min |
| Usability | UX issues | 5-8 | 30-60 min |
| Observation | Real behavior | 3-5 | 2-4 hours |
| Analytics | Scale patterns | All users | Ongoing |
Qualitative Research
Interview Structure
OPENING (5 min):
- Intro & rapport
- Permission to record
- Context setting
CONTEXT (10 min):
- Role and responsibilities
- Day-to-day workflow
- Tools used
DEEP DIVE (20 min):
- "Walk me through [process]..."
- "Tell me about last time [problem]..."
- "What frustrates you most?"
IMPACT (10 min):
- "What happens when [problem]?"
- "How much time/money does it cost?"
FUTURE (10 min):
- "What would ideal look like?"
- "What would you pay for [solution]?"
CLOSING (5 min):
- "Anything else?"
- "Can I follow up?"
Interview Best Practices
- Listen 70%, talk 30%
- Ask "Why?" 5 times
- Avoid leading questions
- Use silence effectively
- Capture quotes verbatim
Quantitative Research
Survey Design
Question Types:
- Rating scale (1-5, 1-10)
- Multiple choice
- Open-ended (limit 1-2)
- Ranking
NPS Question: "How likely are you to recommend [product] to a friend? (0-10)"
Sample Size Calculator
For 95% confidence, 5% margin:
- Population 100 → Sample 80
- Population 500 → Sample 217
- Population 1000 → Sample 278
- Population 10000 → Sample 370
Synthesis
Affinity Mapping
- Write each insight on sticky note
- Group similar insights
- Name each group (theme)
- Rank by frequency/impact
- Extract top 5-10 themes
Persona Template
NAME: [Descriptive name]
ROLE: [Job title, company type]
QUOTE: "[Real quote from research]"
GOALS:
- [Goal 1]
- [Goal 2]
FRUSTRATIONS:
- [Pain 1]
- [Pain 2]
BEHAVIORS:
- [How they work]
- [Tools they use]
NEEDS:
- [Need 1]
- [Need 2]
Journey Map
| Stage | Actions | Emotions | Pain Points | Opportunities |
|---|---|---|---|---|
| Aware | Search | Curious | Hard to find | SEO, content |
| Consider | Compare | Confused | Too many options | Comparison |
| Purchase | Buy | Anxious | Complex checkout | Simplify |
| Use | Onboard | Overwhelmed | Steep learning | Better UX |
Troubleshooting
Yaygın Hatalar & Çözümler
| Hata | Olası Sebep | Çözüm |
|---|---|---|
| Low response | Wrong incentive | $50-100 gift card |
| Surface insights | Leading questions | "Why?" 5x |
| Conflicting data | Mixed segments | Segment analysis |
| No show | Scheduling issues | Calendar hold, reminder |
Debug Checklist
[ ] Research plan documented mi?
[ ] Sample size sufficient mi?
[ ] Questions non-leading mi?
[ ] Recording consent alındı mı?
[ ] Synthesis done within 24h mi?
[ ] Insights actionable mi?
Recovery Procedures
- Low Participation → Increase incentive, new channels
- Conflicting Data → Segment by user type
- Shallow Insights → Follow-up interviews
Learning Outcomes
- Plan effective research studies
- Conduct insightful interviews
- Design valid surveys
- Synthesize research data
- Present actionable insights
Score
Total Score
Based on repository quality metrics
SKILL.mdファイルが含まれている
ライセンスが設定されている
100文字以上の説明がある
GitHub Stars 100以上
1ヶ月以内に更新
10回以上フォークされている
オープンIssueが50未満
プログラミング言語が設定されている
1つ以上のタグが設定されている
Reviews
Reviews coming soon
