| name | usability-testing |
| description | Usability testing methodology - study design, task creation, moderation techniques, analysis, and reporting. Use when planning user tests, writing test scripts, analyzing results, or presenting findings. |
Usability Testing
Most teams skip usability testing because they think it's slow and expensive. It's neither. Five users, one afternoon, and a prototype that doesn't even need to work. That's it. The alternative is shipping something nobody can use and spending 3 months fixing it. Testing is faster.
When to Use This Skill
Auto-loaded by agents:
- usability-researcher - For all research work
Use when you need:
- Test plan for a design or prototype
- Task scenarios for usability testing
- Analysis framework for test results
- Report template for stakeholders
Test Types
| Type | Participants | Best For | Cost |
|---|
| Moderated (in-person) | 5-8 | Complex flows, early concepts, nuanced feedback | High |
| Moderated (remote) | 5-8 | Same as above, broader geography | Medium |
| Unmoderated (remote) | 15-30 | Validation, specific tasks, quantitative metrics | Low |
| Guerrilla | 3-5 | Quick feedback, low-fi prototypes, cafes/hallways | Very low |
| A/B testing | 1000+ | Comparing two specific variations, measuring conversion | Medium (time) |
Writing Good Tasks
Bad: "Find the settings page"
Good: "You want to change your notification preferences so you only get emails about sales. How would you do that?"
Task Principles
- Frame as realistic scenarios, not instructions
- Don't use the same words as the UI labels
- Include context and motivation
- One goal per task
- Write 5-7 tasks per session (30-45 min)
Key Metrics
| Metric | What It Measures | Target |
|---|
| Task success rate | Can users complete the task? | >78% (average benchmark) |
| Time on task | How long does it take? | Varies by complexity |
| Error rate | How many mistakes? | <2 per task |
| SUS score | Overall perceived usability | >68 (average), >80 (good) |
| Task difficulty (SEQ) | Perceived ease | >5 on 7-point scale |
The 5-User Rule
Jakob Nielsen's research shows 5 users find ~85% of usability problems. Run small tests frequently rather than large tests infrequently.
The biggest mistake teams make isn't skipping testing. It's testing to validate instead of testing to learn. If you go in wanting to prove your design works, you'll ignore every signal that it doesn't. Go in wanting to find problems. They're there.
Related Skills
design-heuristics - Expert evaluation (complements user testing)laws-of-ux - Principles for diagnosing why users struggle
