What Is Usability Testing? How to Test Your App or Website With 5 Users
Usability testing means watching real people try to use your product. Five users are enough to find most of the serious problems. Here is what it is, when to do it, and how to run one yourself this week.
Usability testing is watching real people try to complete real tasks with your product, to find where they struggle. You give a person a task, such as "order a cake for Friday", stay quiet, and observe. It is not a survey and not a focus group: you study what people do, not what they say they would do. Testing with about five users is usually enough to find most of the serious problems in one flow.
It is one of the cheapest ways to improve an app or website, and it can be done on a prototype before any code is written.
Why are five users enough?
Research published by the Nielsen Norman Group found that the first few users reveal most of the usability problems, and that after about five the same problems keep repeating. Their estimate is that five users uncover roughly 85% of the problems in the flow being tested.
Two conditions apply:
- The five people should be from one type of user. If you have two very different audiences, such as customers and drivers, test about five of each.
- It is better to run several small rounds than one large one: test with five, fix, then test again.
What are the types of usability testing?
| Type | How it works | Best for |
|---|---|---|
| Moderated | You sit with the person, in the room or on a video call, and watch | Understanding why people struggle |
| Unmoderated | The person completes tasks alone while a tool records the screen | Quick feedback from many people |
| In person | Same room, their own phone if possible | Seeing real behaviour and context |
| Remote | Video call with screen sharing | Reaching users in other cities or countries |
| Prototype testing | Tasks on a clickable design, before development | Catching problems while they are cheap to fix |
| Live product testing | Tasks on the real app or site | Finding out why a released flow underperforms |
When should you run a usability test?
- Before development, on a prototype of the main flow
- Before a redesign, to learn what is actually broken. See how to redesign an app without losing users
- After launch, when analytics show people leaving a flow and you do not know why
- When the team disagrees about a design decision: watch users instead of arguing
How do you run a usability test? Seven steps
1. Choose one flow. Sign up, checkout, booking. Do not test the whole product at once. 2. Write three to five tasks as situations, not instructions. Write "You want to send flowers to your mother on Thursday", not "Tap the flowers category". 3. Recruit five people who match your real users. Not colleagues, and not friends who already know the product. 4. Prepare the product or prototype and a way to record the screen and voice, with the person's permission. 5. Ask them to think aloud, then stay quiet. Do not help and do not explain. If they ask what to do, ask what they would do if you were not there. 6. Take notes on what they do: where they pause, tap the wrong thing, go back or give up. 7. Afterwards, list the problems by how many people hit them and how badly, and fix the top ones first.
What should you measure?
| Measure | What it tells you |
|---|---|
| Task completion | Could they finish at all? |
| Time on task | How much effort it took |
| Errors | Wrong taps, wrong pages, going back |
| Where they hesitated | Unclear labels or layout |
| What they said | The words they use, which should become your labels |
What mistakes ruin a test?
The moment you explain the screen, the test is over. In real life nobody is standing next to the user.
- Leading questions, such as "Was that easy?"
- Helping when the person gets stuck
- Testing with the team or with people who already know the product
- Asking for opinions on colours instead of giving tasks
- Testing too late, when nothing can be changed
- Collecting findings and fixing nothing
Usability testing, UX audit or A/B test?
| Method | What it is | Use it when |
|---|---|---|
| Usability testing | Watching users complete tasks | You need to know why people struggle |
| UX audit | An expert reviews the product against known principles | You want a fast list of likely problems. See what a UX audit is |
| A/B test | Two versions shown to live traffic, compared by numbers | You have enough traffic and a specific change to compare |
They work well together: an audit finds the likely problems, testing confirms them with real people, and an A/B test measures the fix.
How Flowlix uses usability testing
Flowlix Studio is a design and development studio in Ismailia, Egypt. We design products and build them: UI/UX design, websites and web apps in React and Next.js, and mobile apps in Flutter and React Native. When a project includes research, we recommend testing the main flow on a clickable prototype before development starts, so problems are fixed in the design. We work with clients in Egypt, Saudi Arabia, the UAE and Qatar. More on our UI/UX design service.
Frequently asked questions
How much does usability testing cost?
A basic round can cost very little: five people, an hour each, and a small thank-you. Costs rise with recruiting specific users, paid tools and professional analysis.
Can I test a design that is not built yet?
Yes, and it is the best time. A clickable prototype is enough for people to attempt real tasks.
How long does a session take?
Usually 20 to 40 minutes per person for one flow. Longer sessions tire people and produce weaker results.
Do I need special software?
No. A video call with screen sharing and recording is enough to start. Dedicated tools help when you test often or want unmoderated sessions.
Is usability testing the same as user acceptance testing?
No. User acceptance testing checks that the software does what was specified. Usability testing checks whether people can actually use it.
Talk to us
Book a free consultation on WhatsApp. Tell us which flow is underperforming, and we will suggest how to test it and what a fix would involve, with a fixed quote within 48 hours.