Menu

UX

Usability Testing

A practical UX method for identifying usability issues by watching real users attempt real tasks.

How to use usability testing to uncover friction, errors, and failure points so you can improve the experience with confidence.

4 min read

What it is

is a UX method used to observe real users attempting to complete tasks using a product or .

Participants are given realistic scenarios and asked to complete specific tasks while their is observed.

It focuses on identifying issues such as confusion, errors, inefficiencies, and .

Unlike analytics, which shows what users do at scale, shows why they struggle.

The goal is to uncover problems in the experience and understand how to fix them.

Usability testing is useful because it turns assumptions about what works into observed evidence of what actually happens.

When to use it

Use this method when you need to understand how well something actually works.

It is most useful when:

You are testing a new design, prototype, or live product
You want to identify usability issues
You need to validate design decisions
You are improving key user journeys
You want to reduce friction and increase success rates

It is less useful when:

You need large-scale quantitative data
The product is not ready to be tested
Tasks are not clearly defined
Usability testing is often used alongside analytics and interviews to combine behaviour with understanding.

Key takeaway

Use usability testing when the main question is whether people can actually complete the task and where the experience is getting in their way.

How to run it

Set up properly

Be clear on what you are testing, which tasks reveal it, and what would count as a failure. A test without a defined failure condition tends to conclude that everything went fine.

Recruit people who resemble your users in the way that matters for this product. Five of the right people beat twenty of the wrong ones.

Run the method

watches people attempt real tasks. It is the most direct evidence available about whether a design works, and the easiest to undermine by helping.

  1. Give clear tasks based on real scenarios, written in the user's language rather than using your 's labels.
  2. Ask people to , and remind them once if they go quiet. Do not repeatedly, which turns narration into .
  3. Observe rather than outcomes. Somebody who succeeds after three wrong turns has found a problem, not passed a test.
  4. Resist helping. The silence while somebody struggles is uncomfortable and is the most valuable part of the .
  5. Capture successes as well as failures. Knowing what already works protects it from being redesigned.

Watch what people do, not what they say afterwards. Post-task rationalisation is confident, articulate and frequently wrong.

Capture and make sense of it

The value comes from observed difficulty. After the , document:

  • Tasks people failed or completed only with difficulty
  • The specific moment each problem occurred
  • Issues that recurred across participants, which are design faults rather than individual ones
  • What worked, so it survives the next round of changes

Use this to fix specific problems. is diagnostic rather than comparative: it tells you what is broken, not which option is better.

What to look for

Focus on:

Completion: whether the task finished, and at what cost
Errors: mistakes or failed interactions
Time on task: how long it takes to complete actions
Confusion: moments where users hesitate or struggle
Workarounds: unexpected ways users achieve their goals

Where it goes wrong

Most issues come from:

What users do matters more than what they say.

Tasks written in your own labels, which tests recall rather than usability
Helping when somebody struggles, which removes the finding
Recording what people said about the design instead of what they did with it
Reading five sessions as a percentage
Producing findings nobody is accountable for fixing

What you get from it

Done properly, this method gives you:

The specific moments where a task broke down
Problems that recurred across participants, which are design faults not individual ones
Evidence strong enough to settle a design argument
Confirmation of what already works, so it survives the next redesign

Key takeaway

It helps you build experiences that actually work.

Get in touch

If this sounds like something you need, we can help you uncover what is not working and fix it quickly.

No guesswork. No assumptions. Just clear insight you can act on.

FAQ

Common questions

A few practical answers to the questions that usually come up around this method.

What is usability testing in UX?

Usability testing is a method used to observe users completing tasks to identify issues in an experience.

When should you use usability testing?

Use it when testing designs, prototypes, or live products.

How many users do you need?

Typically 5 to 8 users per round is enough to identify most usability issues.

What is the think-aloud method?

It involves asking users to verbalise their thoughts while completing tasks.

Does usability testing improve UX?

Yes. It directly identifies problems that impact usability and success.

Quick take

If you want to see where users struggle, get stuck, or fail to complete tasks, run usability testing.

LET'S WORK TOGETHER

Ready to improve your product?

UX, research and product leadership for teams tackling complex digital services.

Previous feedback

I had a fantastic experience working with Andy. One of his most impressive achievements during our time at NHS HEE was masterminding a deeply complex information architecture for a new platform that brought together a large number of legacy websites.

Will Parkhouse

Senior Content Designer