Guessing what users want is expensive. A single usability test with 5 users reveals 80% of usability problems. Research isn’t a luxury — it’s a cost-saving measure.

Research methods by question

Question	Method	Time	Confidence
What do users need?	Contextual inquiry	2-4 weeks	High
Can users complete this task?	Usability test	1-2 weeks	High
Which design is better?	A/B test	2-4 weeks	Very high
Why are users confused?	Think-aloud protocol	1 week	High
How many users have this problem?	Survey	1-2 weeks	Medium
What do users expect?	Card sorting	1 week	Medium

Usability test script

## Pre-test (5 min)
- Thank participant, explain format
- "There are no wrong answers — we're testing the design, not you."
- Record screen + audio

## Tasks (20 min)
1. "Imagine you just signed up. Find the settings page."
2. "Change your email address to a new one."
3. "You want to invite a teammate. Show me how you'd do that."

## Post-test (5 min)
- "What was your overall impression?"
- "What was the most frustrating part?"
- "If you could change one thing, what would it be?"

Interview protocol

interface InterviewQuestion {
  category: 'context' | 'behavior' | 'motivation' | 'pain-point';
  question: string;
  followUp?: string;
  avoid?: string; // leading words to avoid
}

const questions: InterviewQuestion[] = [
  {
    category: 'context',
    question: "Walk me through your typical workday.",
    followUp: "What tools do you use most frequently?",
  },
  {
    category: 'behavior',
    question: "Tell me about the last time you dealt with [problem].",
    followUp: "What did you try first? What happened next?",
    avoid: "Why did you do it that way?" (sounds accusatory)
  },
  {
    category: 'motivation',
    question: "What would make this easier for you?",
    avoid: "Would you use feature X?" (everyone says yes)
  },
];

Affinity mapping

## Process
1. Write each insight on a sticky note (one insight per note)
2. Group related notes into clusters
3. Name each cluster (emergent theme)
4. Look for patterns across clusters
5. Prioritize by frequency × impact

## Template
### Theme: [Cluster Name]
- Insight 1 (3/5 participants)
- Insight 2 (4/5 participants)
- Insight 3 (2/5 participants)

### Priority: [High/Medium/Low]
### Recommendation: [Action to take]

Research repository

Organize findings for long-term value:

research/
  participants/
    P001-interview.md
    P002-usability-test.md
  themes/
    navigation-pain-points.md
    onboarding-friction.md
  reports/
    2024-q1-usability-audit.md
  artifacts/
    journey-maps/
    personas/

Anti-patterns

Don’t ask “Would you use this?” — watch what they actually do
Don’t test with more than 5 users per round — diminishing returns
Don’t lead participants — “Don’t you think this is confusing?”
Don’t skip the debrief — insights fade fast
Don’t research without action — every study needs a decision it informs

When it triggers

conducting user research
usability testing plan
user interview questions
analyzing user feedback
prioritizing feature requests

UX Research