UX researchers are often faced with the question of how many participants are required to conduct a successful study. Five participants, also known as “magic number 5”, have been considered the traditional number in UX research, but is this number sufficient?

The answer depends on the research goals, the complexity of the product, and the research methodology employed.

5 Participants Is Enough

Some studies (Virzi, 1992; Nielsen & Landauer, 1993; Lewis, 1994) believe that 80% of usability issues can be discovered with the first five participants. Nielsen & Landauer (1993) analyzed 11 different studies and they discovered that there is a probability of finding out 31% of usability problems per participant.

As you can see, after the 3rd participant more than 60% of usability issues were detected, after the 3rd participant 66%, and after the 6th more than 80%.of usability problems were observed.

5 Participants Is Not Enough

In some cases, five participants may be enough to uncover the majority of usability issues. However, in more complex studies, a larger sample size may be necessary to achieve statistically significant results.

Some researchers have challenged the “magic number 5”. One example is Spool and Schroeder's (2001) study where participants were asked to purchase CDs and DVDs on 3 different websites, the results showed that only 35% of usability issues were found after the first five participants.

Scope Versus The Sample

As a UX researcher, it's crucial to determine the appropriate sample size for a study to ensure that the results are statistically significant and accurate. In my experience, 5 participants are enough to conduct a usability study if the scope is smaller, for example, if your goal is to test one process like the “set up your profile” wizard.

For more complex tasks or whole products, results are more complete and precise with a bigger sample. Even if the scope is large you can discover some general usability issues which can be helpful in making broad product decisions and are absolutely worth discussion with stakeholders.

One major benefit of using a small sample size is cost-effectiveness. Recruiting and compensating participants can be expensive, and a smaller sample size can help reduce costs. Additionally, a smaller sample size allows for a more in-depth analysis of each participant's behavior.

On the other hand, a larger sample size can provide more generalizable results, which can be helpful in making broad product decisions. However, a larger sample size can also lead to more complex data analysis and requires more resources.

In conclusion, the question of whether five participants are enough depends on the research goals.