How to Run a Usability Test

Usability testing is one of the most valuable activities in product design and one of the most often skipped. Teams worry it's expensive, slow, or requires specialist skills. In reality, a basic usability test can be run in a few days with a handful of participants and a prototype, and it will almost always reveal problems the team didn't see.

This guide walks through the process step by step.

What usability testing is — and isn't#

In a usability test, you watch real people try to complete realistic tasks with your product or prototype. The goal is to find where they struggle, misunderstand, or fail, so you can fix those issues.

It's not:

A preference survey. You're observing behavior, not asking what people like.
A focus group. Sessions are one-on-one, so participants don't influence each other.
A demo. You don't explain the interface; you watch people figure it out.

Step 1: Define what you want to learn#

Start with clear goals. Good goals are specific:

"Can new users create and share their first project without help?"
"Do people understand the difference between the two pricing plans?"
"Can returning customers find and download past invoices?"

Vague goals like "test the new design" produce vague findings. Two to four focused goals per study is usually enough.

Step 2: Choose your method#

The main choices are:

Moderated vs. unmoderated. Moderated sessions, in person or remote, let you ask follow-up questions and understand why people struggle. Unmoderated tests are faster and cheaper but provide less depth.
In person vs. remote. Remote testing is easier to schedule and reaches more diverse participants. In-person testing is useful for physical products or contexts where observing the environment matters.

For most design questions, remote moderated sessions offer a good balance of depth and practicality.

Step 3: Recruit the right participants#

Participants should resemble your actual or intended users. Testing a professional accounting tool with random consumers will produce misleading results.

Define key characteristics: role, experience level, behaviors, and needs.
Use a short screener survey to filter candidates.
Exclude people too close to the product, like employees or their friends.
Offer fair compensation for their time.

How many participants? A widely cited guideline is that around five participants per distinct user group reveal most of the common usability problems in a design. It's better to run several small rounds of testing, fixing issues between them, than one large study.

Step 4: Write realistic tasks#

Tasks are the heart of a usability test. Good tasks:

Reflect real goals: "You need to send last month's invoice to your accountant."
Avoid giving away the answer: don't use the exact words from the interface. If the button says "Export," don't say "Export the report."
Have a clear end point, so you know whether the task succeeded.
Include context: a short scenario helps participants behave naturally.

Order tasks logically, starting with something simple to build confidence. Plan for five to eight tasks in a 45 to 60-minute session.

Step 5: Prepare the prototype and script#

Make sure the prototype supports every task path, including likely wrong turns.
Use realistic content rather than placeholder text.
Write a script with an introduction, background questions, tasks, and wrap-up questions.
Run a pilot session with a colleague to catch broken links and confusing tasks.

Step 6: Moderate without leading#

Moderating is a skill, but a few principles go a long way:

Set expectations. Explain that you're testing the design, not them, and that there are no wrong answers.
Ask them to think aloud. Encourage participants to say what they're looking at, thinking, and expecting.
Stay neutral. Avoid reacting to success or failure. Don't say "great" when they find something.
Don't help. If they're stuck, ask "What would you do if you were on your own?" Only intervene if they're completely blocked, and note that the task failed.
Ask open questions. "What are you thinking now?" or "What did you expect to happen?" rather than "Was that confusing?"
Embrace silence. Pauses often lead participants to explain their thinking.

Step 7: Observe and take notes#

Record sessions with permission, but also take notes during each one. Capture:

Whether each task succeeded, partially succeeded, or failed.
Where participants hesitated, backtracked, or made errors.
Direct quotes that illustrate problems.
Unexpected behaviors and workarounds.

Invite team members — designers, developers, product managers — to observe live. Watching real users struggle is far more persuasive than reading a report.

Step 8: Analyze findings#

After all sessions:

Collect observations from notes and recordings.
Group similar issues across participants.
Rate severity based on how many people were affected and how badly — for example, critical, serious, or minor.
Identify root causes. A problem might show up as a missed button, but the real cause may be unclear labeling or poor hierarchy.

Focus on patterns rather than individual opinions. One participant disliking a color isn't a finding; three participants failing to find the same feature is.

Keep the report short and actionable:

Summarize goals, method, and participants.
List key findings ordered by severity.
Support each finding with evidence: quotes, clips, or screenshots.
Provide clear, specific recommendations.
Highlight what worked well, too.

Short video clips of real users are often the most powerful way to communicate problems to stakeholders.

Step 10: Fix and test again#

Usability testing is most effective as a cycle. Fix the most serious issues, then test again to confirm the fixes work and to find the next layer of problems.

Common mistakes to avoid#

Testing with the wrong participants.
Writing tasks that reveal the answer.
Helping participants when they struggle.
Asking people what they'd like instead of watching what they do.
Running one big study instead of several small ones.
Producing long reports nobody reads.

Small tests, big insights#

Usability testing doesn't need to be expensive or complicated. A clear goal, the right participants, realistic tasks, and careful, neutral observation will uncover problems that no amount of internal debate can find.

Make it a regular habit — even a few sessions every few weeks — and your products will improve steadily, based on evidence rather than assumptions.

How to Run a Usability Test: A Step-by-Step Guide