It's important to plan how many participants you need to recruit to meet your research goals with reliable results.
Whether you're testing a navigation structure, evaluating a prototype, conducting interviews, or gathering survey responses, the right sample size helps you identify meaningful patterns and make informed decisions.
Use this guide to plan participant numbers for your Optimal studies, understand when you need a larger sample, and decide when you have enough data to move forward.
Quick Guide: Recommended Participant Numbers
These ranges are starting points, not strict requirements. The right sample size depends on whether you're exploring behavior, identifying usability issues, or measuring performance.
Research Method | Recommended Starting Point | Best For |
Usability Testing (Mixed Methods) | 5–8 participants for qualitative insights; 20+ when measuring quantitative outcomes | Combining behavioral metrics with qualitative feedback to understand both what users do and why |
Moderated Interviews | 5–10 for an initial exploratory round; continue until you understand the key patterns | Observing behavior and understanding usability issues in depth |
AI Interviewer | 10-15 for an initial exploratory round; 25-50 for exploring complex questions | Conducting in-depth interviews at scale and exploring participant perspectives with AI-led follow-up questions |
Prototype Testing | 5–8 participants for qualitative testing; 20+ for quantitative evaluation | Identifying usability issues and evaluating task completion in interactive prototypes |
Live Site Testing | 5–8 participants for qualitative testing; 20+ for quantitative evaluation | Understanding how people use a live website and identifying usability issues in real-world journeys |
Surveys | Varies | Measuring opinions, preferences, and reported behaviour |
Card Sorting | 15–30 for exploratory research; 30–50+ for quantitative analysis | Understanding how people group and label information |
Tree Testing | 15–30 for exploratory research; 30–50+ for quantitative analysis | Evaluating navigation and information findability |
First-click Testing | 50–100 participants | Assessing whether people know where to start a task |
Qualitative Insights | Depends on the number and depth of sessions or observations collected | Identifying themes across interviews, usability sessions, and observations |
These recommendations combine draw on common UX research practices and guidance, but are not intended to be guarantees of statistical significance.
Usability Testing
Usability testing helps you understand whether people can use a website, product, or prototype to complete their goals. The number of participants you need depends on whether you're looking for problems to fix or measuring performance.
Qualitative usability testing: Start with five participants for a focused test with a relatively consistent audience. This can reveal many common usability problems.
Quantitative usability testing: Aim for at least 20 participants as a starting point when measuring task success, time on task, or other performance metrics. Larger samples may be needed for precise estimates or comparisons.
Complex experiences: Consider additional participants when testing multiple flows, device types, or user groups.
Iterative testing: Test, improve the design, and run another round with new participants to check whether the changes have addressed the problems.
Moderated Interviews
In qualitative research, such as in-depth interviews or focus groups, the emphasis is on gathering detailed insights rather than achieving statistical significance. Here’s what to keep in mind:
Early discovery: Start with five to ten interviews to identify recurring experiences, motivations, and pain points.
Different audience groups: Recruit across the relevant groups and analyze their experiences separately where appropriate.
Complex or sensitive topics: Allow time for deeper follow-up and consider whether a trained human moderator is best suited to the subject.
Further rounds: Continue recruiting when new interviews reveal important perspectives or challenge your existing understanding.
AI Interviews
AI Interviews can help you scale your research, collect and analyze more interviews efficiently. However, the number of participants you need still depends on the research question, the diversity of the audience, and the depth of evidence required. We recommend 10-15 for an initial exploratory round or up to 50 for exploring more complex questions.
Prototype Testing
The recommended number of participants for prototype testing is 20 or more. Here are a few suggested guidelines:
Initial Usability Feedback: With 20 or more participants, you can start to identify major usability issues. This sample size is generally sufficient to uncover common user challenges and validate design decisions.
Iterative Testing Benefits: If significant changes are made to the prototype after the first round, an additional round of testing with 10-15 new participants can provide validation for the updated design.
Complex Prototypes: For more complex products with numerous interactions, a larger sample size can provide more comprehensive feedback and help identify edge cases.
Live Site Testing
Live site testing helps you understand how people interact with a live website or digital experience, using real pages and functionality rather than a static or interactive prototype.
Qualitative testing: Start with five to eight participants to identify usability issues, observe behaviour, and understand where people encounter difficulties.
Quantitative evaluation: Aim for 20 or more participants as a starting point when measuring task success, completion rates, or other performance metrics. Larger samples may be needed for greater precision or to compare groups.
Complex journeys: Recruit additional participants when testing multiple tasks, user journeys, devices, or audience segments.
Iterative improvements: Address the issues you identify, then run another round of testing to evaluate whether your changes improve the experience.
When analyzing results, look beyond whether participants complete a task. Consider where they hesitate, take unexpected paths, or encounter errors, and use participant feedback to understand why these issues occur.
Surveys
Surveys and qualitative research are research methods with a wide range of use cases. The sample size for surveys can range from a few participants to several thousand participants. For general surveys seeking broad insights, larger sample sizes help to ensure statistical reliability and generalizability.
Card Sorting
Card sorting helps you understand how people group information and what labels they expect to find. It is particularly useful when designing or improving website navigation, product menus, and information architecture.
Exploratory card sorting: Start with 15–30 participants to identify recurring groupings and understand how people organise information.
Quantitative card sorting: Aim for 30–50 or more when you need greater confidence in common groupings, want to compare audience segments, or are making significant information architecture decisions.
Comparing groups: Recruit enough participants in each segment to make meaningful comparisons.
Tree Testing
Tree testing evaluates whether people can find information within a proposed navigation structure, without the influence of visual design.
Early validation: Around 15–30 participants can provide useful directional feedback when you're refining a proposed structure.
More confident validation: Aim for 30–50 participants when assessing findability across a range of tasks.
Comparing structures or audience groups: Plan for a sufficient sample in each version or segment, particularly when results will inform a major navigation decision.
Look beyond the overall success rate. Examine directness, task completion, incorrect paths, and the tasks that cause the most difficulty. Individual tasks may receive fewer completed responses than the total participant count, so check your data at the task level.
First-click Testing
First-click testing is one of the quickest and most straightforward UX research methods to set up and quick and easy for participants to complete. First-click testing helps you understand where people expect to click first when completing a task on a design, screenshot, or interface.
Evaluating designs: We recommend recruiting for 50-100 participants for your studies. For comparing designs, you can aim for 25–50 participants per version (50-100 total) as an initial benchmark, with larger samples when you need more confidence in differences.
High-stakes decisions: Recruit more participants and consider whether the results need to support quantitative conclusions.
Make sure participants receive a clear, realistic task. A correct first click can indicate that a design communicates the right starting point, but it doesn't prove that someone can complete the entire journey. When comparing designs, keep the task wording and target audience consistent so you're comparing like with like.
Qualitative Insights
Optimal's Qualitative Insights tool help you organize observations, identify themes, and connect evidence across research sessions. The number of participants needed depends on the type of research being analyzed.
For example:
Focused usability research: Five to eight sessions can be a useful initial round for a specific task or journey.
Broader discovery research: Ten to fifteen or more interviews or sessions may be needed when exploring a complex problem or multiple perspectives.
Research across several groups: Collect enough evidence from each relevant audience to understand meaningful differences.
Ongoing research programs: Analyze findings as you go, identify gaps, and recruit further participants where needed.
Prioritize the depth, relevance, and quality of the evidence. More sessions can help reveal additional perspectives, but they won't compensate for a poorly defined research question or an audience that doesn't reflect your users.
When do you have enough participants?
Use your research goals and the quality of the evidence to decide when to stop recruiting.
You may have enough participants when:
You're seeing consistent patterns that answer your research questions.
You've identified the most important usability issues and understand their impact.
Your quantitative results have sufficient precision for the decision you're making.
You have enough responses within each audience group to make the comparisons you planned.
Additional participants are no longer revealing substantially different insights — while remaining alert to underrepresented perspectives.
If your findings are inconsistent, your target audience is diverse, or your decisions carry significant risk, recruit more participants or run another round of research.