How many testers do you need for UX Research?

Determining how many respondents you need to conduct a study with to get reliable results can be tricky. In this article, we explain the difference in numbers for qualitative and quantitative UX research and provide a guideline for the number of respondents for each type of study. 

  1. Number of test users in qualitative UX research
  2. Number of test users in quantitative UX research
  3. Number of test users in a consumer needs study
  4. Number of test users in persona studies
  5. Number of test users in target group research
  6. Number of test users in customer journey research
  7. Number of test users in concept testing
  8. Number of test users in prototype research (user testing)
  9. Number of test users in navigation research (card sorting and tree testing)
  10. Number of test users in usability research
  11. Number of test users in UX Benchmarking

Difficulties with recruiting test participants? We are happy to help you out.

Book a demo
 

Number of test users in qualitative UX research

In qualitative UX research, the focus is on gaining deep understanding and insight through individual interviews, group discussions or observations. The numbers in qualitative UX research tend to be smaller compared to quantitative research. A characteristic of qualitative UX research is that you want to continue conducting research to the point where you do not bring in any or very limited new information.

The numbers here are often between 5 and 20 respondents per user group and research. While this number may seem limited, it often provides enough information to gain important insights into user needs, preferences and behaviours.
 

Number of test users in quantitative UX research

Quantitative UX research, on the other hand, focuses on collecting data on a larger scale to gain generalisable insights. Questionnaires are often used for this purpose. The number of respondents required for quantitative research depends on several factors, including the desired reliability and accuracy of the results, the complexity of the research question and the population size.

The respondent numbers for reliable research results in quantitative UX research are often between 100 to 200 respondents.
 

Number of test users in a consumer needs study

In a consumer needs study, it is important to have a representative sample that reflects the diversity of the potential user population. A good guideline is to include at least 15 to 30 respondents in qualitative interviews or group discussions. These numbers allow researchers to identify common patterns and trends in needs.

Numbers of test users in persona studies

Creating user personas involves creating fictional characters representing key user groups. The purpose of this research is to understand the behaviour, goals and demographics of different user groups. Generally, a minimum of 15 to 20 respondents is sufficient to create reliable personas.
 

Number of test users in target group research

Target group research looks specifically at the characteristics and preferences of the product's primary target group. Here, quantitative methods such as online surveys can be used. The number of respondents required can vary, but generally a sample of at least 100 to 200 respondents is sufficient to gain reliable insights about the target group. In addition, interviews or focus groups can be conducted, for which often between 10 and 20 respondents are recruited.
 

Number of test users in customer journey research

Customer journey research focuses on understanding users' interaction experiences throughout the process, from awareness to purchase and use of a product or service. This research can include both qualitative and quantitative methods, such as interviews, focus groups, observations and questionnaires. The number of respondents can vary, but typically a minimum of 15 to 20 respondents per phase of the customer journey is recommended to obtain sufficient insights.
 

Number of test users in concept testing

Concept research or concept testing involves testing different concepts or design alternatives to determine which are the most promising. For concept research, quantitative methods such as online surveys can be used. A sample of 100 to 200 respondents is then often sufficient to make a reliable comparison between concepts.

It is also possible to use qualitative methods such as interviews and focus groups for concept testing. In the case of a focus group, 6 to 10 respondents are often invited. In the case of interviews, the numbers are often between 10 and 20.
 

Number of test users in prototype research (user testing)

To test a prototype, user testing is used. This phase is an important part of the design process, as changes can still be made before development takes place. The number of respondents for prototype user testing can vary, but generally 5 to 10 respondents per user group are recommended. This number allows researchers to identify key usability issues and iterations before the product is launched.
 

Number of test users in navigation research (card sorts and tree tests)

In navigation research, card sorting and tree testing are often used. For card sorts, a sample size of 15 to 20 respondents is sufficient, while for tree tests, 10 to 15 respondents per test often yield reliable results.
 

Number of test users in usability research

Usability research focuses on identifying and solving specific usability problems in an existing product. Usability research makes use of user tests (also called UX tests or user testing). For reliable results, it is recommended to test with 5 to 10 respondents per user group and device type.
 

Number of test users in UX Benchmarking

UX Benchmarking compares the user experience of a product with that of competitors or previous versions of the same product. The number of respondents required here depends heavily on the research methods chosen. Do you use a standardised questionnaire? Then it is recommended to have between 100 and 200 respondents per device type. For a qualitative comparison, 20 respondents is often enough.