User research methods guide
How to choose the right research method
Choosing the wrong research method wastes time and produces misleading data. Here is how to choose between qualitative and quantitative research, and which specific methods to use for each type of product question.
The fundamental distinction: generative vs evaluative
Before choosing a method, answer one question: do you know what to build? Everything follows from there.
Generative research (discovery)
You do not know what to buildYou are trying to understand the user's world, problems, and unmet needs. The output is insight — not a validated design. Methods: user interviews, contextual inquiry, diary studies.
Evaluative research (validation)
You have built somethingYou have built or designed something and want to know if it works. Methods: usability testing, A/B testing, surveys, analytics review.
The most common research mistake: running evaluative research on the wrong thing, because generative research was skipped. You can perfectly validate a solution to a problem nobody has.
Research method by question type
Start with the question you need to answer. The method follows from that — not the other way around.
Question
Why do users do X?
User Interviews
Qualitative1-on-1 conversations with 5–8 users. Ask about behavior, not opinions. Open-ended questions only. Probe for the why behind the why — every answer surfaces another question worth asking.
Question
Do users have this problem?
50+ responses for statistical reliability. Closed-ended questions for quantification, open-ended for texture. Best used to validate hypotheses formed in interviews — not to replace them.
Question
Can users complete this task?
5–8 participants. Give them a realistic task, then observe without helping. Note where they get stuck, what they misunderstand, and what behavior they expect from the interface.
Question
Which version performs better?
Requires sufficient traffic — typically 1,000+ sessions per variant per week for statistical significance. Tests actual behavior, not stated preference. Gold standard for conversion decisions.
Question
What are users actually doing?
Session recordings (Hotjar, FullStory), funnel analysis, click heatmaps. Tells you what is happening, not why. Always pair analytics findings with qualitative research to understand cause.
The five common research mistakes
Most research problems are not about method selection — they are about how the research is conducted. These five mistakes account for the majority of research that misleads product decisions.
1
Asking leading questions
'How useful do you find this feature?' versus 'Walk me through how you would use this.' The first anchors on usefulness before the participant has formed a view.
2
Testing with the wrong users
Participants who are not representative of your target users produce misleading findings. Recruiting criteria matter as much as the research design itself.
3
Confusing research with validation
Going into interviews already knowing the answer and looking for confirmation. If you are not open to being surprised, you are not doing research.
4
Under-recruiting
3 interviews is not enough for pattern recognition. 6–8 is the minimum for qualitative themes to emerge reliably across participants.
5
Skipping synthesis
Raw interview notes are not findings. The research value is in synthesis — identifying what patterns emerge across participants, not cataloguing individual quotes.
The research plan template
A research plan forces clarity before you recruit. If you cannot answer these five questions, you are not ready to run the study.
What you are trying to learn. One sentence. If you cannot write this, the study is not ready.
Which method you are using and why this method answers this question better than alternatives.
Who you are recruiting, how many, and what the recruiting criteria are. Screener questions go here.
Recruiting window, session dates, synthesis period, and when insights will be shared with the team.
What the team will receive at the end — insights report, affinity diagram, recommendation deck, or a topline summary. Name it before you start.
Build research into your practice
Explore the UX Designer career path
User research is a core UX skill. The UX career path covers research methods, information architecture, prototyping, and portfolio development — end to end.
Explore UX career