A better question.
A better method.
Choose a research approach that fits what you need to learn. Leave with a reasoned starting point and a worksheet you can use with your team.
Start with the decision.
The method follows the question. Your answers will reveal a starting point, its limits, and the preparation it needs.
For example
“Where does our signup flow break?” with a prototype and access to participants points toward usability testing.
Try this example →How this guide makes a recommendation
These are editorial starting rules for five common questions, not a complete catalogue or a score of method quality. Combine methods when a decision needs more than one kind of evidence.
- Needs and experiences → interviews.
- Task friction + a testable flow → usability testing.
- Content groupings → card sorting; an existing hierarchy → tree testing.
- Known, self-reported experiences → survey design, with a defensible sampling plan.
- Causal impact + a live product → experiment planning, subject to feasibility checks.
- Missing participants or materials → resolve the prerequisites in a research plan first. Live experiments have separate population and measurement requirements.
For navigation questions, a prototype or live product is assumed to include a hierarchy you can extract. If it does not, choose “An idea or content list.” For complex or high-stakes decisions, review the plan with an experienced researcher.
Grounded in research practice
The selection rules are this wiki’s editorial synthesis. Method details and limitations are linked in each recommendation.
- When to Use Which User-Experience Research Methods — Nielsen Norman Group
- Plan a round of user research — GOV.UK Service Manual