User Testing on a Budget: Five People, One Afternoon, Big Insights

Most website problems are invisible to the people who built the site. You know where everything is, what every label means and how the booking form works. Your customers do not. User testing means watching real people try to use your website, and it is one of the fastest ways to find out what is confusing them.

Many business owners assume user testing needs a lab, special software and a big budget. It does not. With a few participants, a laptop or phone and an afternoon, you can uncover the issues that cost you enquiries. Here is how.

Why a small number of testers is enough

You do not need dozens of people to learn something useful. Usability researchers, notably the Nielsen Norman Group, have long argued that testing with around five users reveals many of the most common problems, and that several small rounds of testing are better than one large study.

The logic is simple: serious usability problems tend to trip up most people. After watching a handful of participants struggle at the same point, you have a clear signal. Fix it, then test again with a few new people.

Recruit the right people

Testers should resemble your real customers as closely as possible. Friends and family can help in a pinch, but they may know your business too well or be too polite. Better options include:

• Recent customers willing to spend half an hour, perhaps for a small voucher or thank-you gift
• People who match your target audience, such as parents for a tuition centre or homeowners for a renovation business
• Colleagues from other departments who have never worked on the website
• Remote participants through online testing services, if your budget allows

Avoid testing only with people who are very comfortable with technology if many of your customers are not. Include a mix of ages and devices where possible, especially phones.

Write realistic tasks

A user test is built around tasks: realistic things a customer might try to do on your website. Good tasks are specific and do not reveal the answer. For example:

1. "You need your aircon serviced next week. Find out how much it would cost and how to book."
2. "You are thinking of enrolling your child in a Year 5 maths class. Find a class that fits Saturday mornings."
3. "You bought a product last week and want to return it. Find out how."

Avoid tasks like "Click the Services menu and find the pricing page", which tell people exactly where to go. Prepare three to five tasks that cover your most important customer journeys.

Key takeaway: the most valuable moments in a user test are when someone hesitates, backtracks or says "Hmm". Those moments show exactly where your website is failing customers.

Run the session without leading

During each session, your job is to observe, not to help. A simple approach:

• Explain that you are testing the website, not the person, and there are no wrong answers
• Ask them to think aloud, saying what they are looking at and thinking
• Give one task at a time and let them work through it
• Stay quiet as much as possible, even when they struggle
• If they ask for help, respond with a question such as "What would you do if I were not here?"
• Take notes on where they hesitate, what they say and whether they complete the task

With permission, record the screen and audio so you can review later or share clips with your team. Free video call tools with screen sharing work well for remote sessions.

Turn observations into fixes

After the sessions, gather your notes and look for patterns. List each problem, how many testers ran into it and how severe it was. Then prioritise:

• Critical: problems that stopped people completing an important task
• Serious: problems that caused significant confusion or delay
• Minor: small annoyances that did not block progress

Fix the critical issues first. Many fixes are quick: clearer labels, more visible buttons, better placement of prices or simpler forms. After making changes, test again with a few new people to confirm the problems are gone. Our post on how to critique your own UI before paying for a redesign offers a self-review checklist to use between testing rounds.

Make testing a habit

User testing is most valuable when it happens regularly, not just once. Consider a short round of testing:

• Before launching a new website or major page
• After a redesign, to catch new problems
• When analytics show a sudden drop in conversions
• Every few months on your most important pages

Even testing wireframes or prototypes before development is worthwhile, because fixing problems at that stage is cheaper. Our guide to wireframing for beginners explains how to build simple clickable prototypes for testing.

FAQ

How long should a user test session be?

Around 20 to 45 minutes is typical. Long sessions tire participants and produce diminishing returns.

Should I test on phones or computers?

Test on the devices your customers use most. For many Malaysian businesses, that means phones first.

Can I test my competitors' websites?

Yes. Watching people use competitor sites can reveal what customers expect and where competitors fall short.

User testing does not need a big budget, only curiosity and a willingness to watch people struggle without stepping in. Recruit a few real customers, give realistic tasks, observe carefully and fix what you find. Few activities improve a website as quickly.