UX Research

Instructions

Research Method Selection — Match the method to the question being asked:

• Generative (what do users need?): Interviews, diary studies, contextual inquiry, ethnographic observation
• Evaluative (does this design work?): Usability testing (moderated and unmoderated), cognitive walkthroughs, heuristic evaluation
• Quantitative (how much / how many?): A/B testing, surveys, funnel analysis, session recording analysis
• Hybrid (structure and behavior): Card sorting (open/closed), tree testing, first-click testing

Choose generative methods early in discovery, evaluative methods during design iteration, and quantitative methods for validation and optimization.

Cognitive Psychology Principles — Design with the brain, not against it:

• Hick’s Law — Decision time increases logarithmically with the number of choices. Reduce options, use progressive disclosure, and set smart defaults.
• Fitts’s Law — Time to reach a target depends on distance and size. Make primary actions large and close to the user’s current focus.
• Miller’s Law — Working memory holds roughly 5–9 items. Chunk information into groups and keep navigation tiers shallow.
• Cognitive load — Minimize extraneous load (decorative clutter, unnecessary steps) and maximize germane load (meaningful learning cues).
• Mental model mapping — Surface how users conceptualize a task before designing flows; card sorting and think-aloud protocols reveal gaps.

Synthesis — Transform raw data into actionable insights:

• Build affinity diagrams to cluster observations into themes
• Create journey maps showing actions, thoughts, emotions, and pain points across phases
• Produce experience maps for cross-channel understanding
• Tag every finding with an evidence tier: Observed (direct data), Inferred (pattern across data points), Assumed (hypothesis needing validation)
• Prioritize with an impact × frequency matrix: high-impact + high-frequency issues get addressed first

Nielsen’s 10 Usability Heuristics — Use as a structured evaluation framework:

1. Visibility of system status — Keep users informed with timely feedback
2. Match between system and the real world — Use familiar language and conventions
3. User control and freedom — Support undo, redo, and easy exits
4. Consistency and standards — Follow platform and industry conventions
5. Error prevention — Design constraints and confirmations that prevent mistakes
6. Recognition rather than recall — Make options and actions visible, not memorized
7. Flexibility and efficiency of use — Provide accelerators for expert users
8. Aesthetic and minimalist design — Remove information that competes with relevant content
9. Help users recognize, diagnose, and recover from errors — Plain-language messages with constructive next steps
10. Help and documentation — Searchable, task-focused, concise

Best Practices:

• Recruit real users, never team members or stakeholders as proxies
• Run usability tests with a minimum of 5 participants per segment to surface ~85% of major issues
• Record sessions with informed consent — video for moderated, screen capture for unmoderated
• Separate observation from interpretation — capture what happened before theorizing why
• Frame findings as: “We observed [behavior], which suggests [insight], and we recommend [action]”

Checklist before delivering research:

• [ ] Research plan reviewed with stakeholders
• [ ] Participant screener validated against target segments
• [ ] Discussion guide piloted with at least one dry run
• [ ] All findings tagged with evidence tier
• [ ] Insight report includes severity ratings (critical / major / minor / cosmetic)
• [ ] Recommendations linked to specific observations