kienbui1995/user-research

User Research

When to Use

When you need to understand user needs, validate a concept, or test whether a design works before or after building.

Core Jobs

1. Choose the Research Method

| Method | Use When | Output | |--------|----------|--------| | User interviews | Understand problems, motivations, context | Qualitative insights | | Usability testing | Test if design works | Issues + severity | | Survey | Quantify attitudes at scale | Stats + verbatims | | Diary study | Understand longitudinal behavior | Behavioral patterns | | Card sorting | Information architecture | Navigation structure |

2. Write the Discussion Guide (Interviews)

Structure:

• Intro (5 min): purpose, permission to record, "no wrong answers"
• Warm-up (5 min): their role, context, how they currently solve the problem
• Core questions (30 min): open-ended, one topic at a time
• Concept test (optional, 10 min): show prototype, ask to think aloud
• Wrap-up (5 min): anything else they'd like to share, referrals

Rules:

• Start broad ("tell me about your workflow") before specific
• Never lead: "Do you find it confusing?" → "What do you think of this?"
• Silence is OK — let them fill it

3. Run the Session

• Recruit 5–8 participants per segment (enough to see patterns)
• Two people: one facilitates, one takes notes
• Record (with permission)
• Note quotes verbatim, not paraphrases

4. Synthesize Findings

• Affinity mapping: group observations by theme
• Write insights: "Users who X tend to Y because Z"
• Don't report "5 users said X" — report what it means

Key Outputs

• Research plan (goals, methods, recruit criteria)
• Discussion guide / test script
• Synthesis report with actionable insights
• Recommendations linked to design or product decisions

Anti-Patterns

• Research after building (too late to change)
• Leading questions that confirm existing beliefs
• Recruiting friends/coworkers (not real users)
• Reporting what users said without interpreting what it means