How to run a five-user usability test this week

Why five

The argument for small tests is old and still holds: the first few participants reveal most of the serious problems in a flow, and each additional person mostly repeats what you have already seen. Five people, one afternoon, one task. Then fix what broke and run it again. Three rounds of five teach you far more than one round of fifteen.

The steps

1. Pick one task

   Choose a task that matters to the business and that a new customer would genuinely attempt, such as “Set up your first invoice and send it to a client.” Write it as a goal, not as instructions.

2. Recruit five people

   Find people who match your customers but have not used this flow. Existing customers who never tried the feature work well. Colleagues from other teams are a last resort; they know too much.

3. Write a short script

   A two-minute welcome, a reminder that you are testing the product rather than them, a request to think aloud, then the task. End with two questions: what was hardest, and what did you expect to happen that didn’t?

4. Run the sessions

   Thirty minutes each, in person or on a video call with screen sharing. When a participant gets stuck, count to ten before you say anything, and then ask what they are looking for rather than showing them.

5. Write it up the same day

   List every problem you saw, how many of the five hit it and how badly. Rank by frequency and severity. Anything that stopped two or more people goes to the top.

6. Fix and repeat

   Change the worst two or three things, then test with five new people. The second round tells you whether the fix worked and shows you the problems the first ones were hiding.

What to avoid

Don’t ask people whether they would use a feature; people are poor predictors of their own behavior. Don’t test five different tasks in thirty minutes. And don’t argue with what you saw. If three of five people missed the button, the button is hidden, whatever the design file says.