To usability test your website, ask five people who match your real customers to complete a few everyday tasks on your site, like finding a price or making an enquiry, while they think out loud. You stay quiet, watch where they hesitate, and write down what they say. One afternoon of this will usually show you the biggest reasons visitors aren't buying or getting in touch.
That's it. No special software, no lab, no agency required. The hard part isn't the method. It's resisting the urge to explain your own website while someone struggles with it.
Below is a simple, owner-friendly way to run your first test, including a script you can copy, example tasks, and how to decide what to fix first.
What is usability testing, in plain English?
Usability testing means watching real people use your website to do something they'd actually want to do. You're not asking for opinions on your colours or logo. You're observing behaviour: where they click, where they pause, what they misread and where they give up.
This matters because what people say and what they do are often different. A customer might tell you your site "looks nice", then take two minutes to find your opening hours. Analytics can tell you that people leave a page. Watching someone use it tells you why.
How many people do you need to test with?
Fewer than you think. Five is a good number for a first round. In a widely cited article, Jakob Nielsen of Nielsen Norman Group argues that testing with about five users uncovers most of the usability problems in a design, and that you'll learn more from several small rounds than one big one.
In practice, by the third or fourth person you'll start hearing the same problems again. That's your signal that you've found something real. If you serve two very different groups (say, retail shoppers and corporate buyers), test a few people from each.
How to run a usability test on your website, step by step
Pick one goal. What do you most want visitors to do? Buy, book a call, send an enquiry, sign up? Test around that, not your whole site.
Write three to five realistic tasks. Phrase them as situations, not instructions. Not "Click on Pricing", but "You're comparing a few options and want to know roughly what this would cost you. Find out." (More examples below.)
Find five people who match your customers. Recent customers, people who enquired but didn't buy, or people in your target group. Avoid your staff, family and close friends. They know too much and they're too kind.
Choose where to run it. In person works well, but a video call with screen sharing is just as good and easier to schedule. Ask them to use the device your customers mostly use. For many Singapore consumer brands, that's a phone.
Run each session for 20 to 30 minutes. Use the script in the next section. Ask them to think out loud as they go.
Stay quiet and take notes. Write down what they did, what they said, and where they hesitated. Exact quotes are gold.
Thank them properly. A small voucher or a discount on their next order is a fair thank-you for their time.
Review all five sessions together. Look for patterns, not one-off comments. Then fix the top issues before testing again.
A simple usability test script you can copy
Read this out at the start of every session, so each person gets the same experience.
"Thanks so much for helping. I'm going to ask you to try a few things on a website. We're testing the website, not you, so there are no wrong answers. If something is confusing, that's exactly what we need to know.
As you go, please say out loud what you're thinking: what you're looking at, what you expect to happen, and anything that surprises you. I might stay quiet while you work. That's on purpose, because I want to see how the site works without me there to help.
You can't hurt my feelings. I didn't design this to be perfect. Ready?"
Then give one task at a time. When they finish (or give up), ask two short follow-ups:
"How did that feel?"
"Was there anything you expected to see but didn't?"
At the very end, ask: "If you could change one thing about this website, what would it be?" and "Is there anything that would stop you buying or enquiring here?"
If you'd like more questions for the end of the session, our list of customer interview questions that get honest answers works well here.
Example usability test tasks for a small business website
Good tasks sound like something a real customer would actually do. Adapt these to your business:
"You've heard about this business from a friend. Spend a minute on the homepage, then tell me in your own words what they do and who it's for."
"You'd like to know roughly how much this would cost before contacting anyone. Find out."
"You want to know if they deliver to your area and how long it takes. Find out."
"You're ready to ask a question about a specific service. Get in touch the way you normally would."
"Find a product you'd consider buying, add it to your cart, and go as far as the payment step. Stop there."
The first task is the most revealing. If people can't explain what you do after a minute on your homepage, nothing else on the site will work as hard as it should.
What to watch for during the test
Your notes should focus on moments, not overall impressions. Look out for:
Hesitation: long pauses, scrolling up and down, hovering without clicking.
Wrong turns: clicking a menu item, then going back.
Misreading: "Oh, I thought this was only for companies."
Missing information: "I'd want to know the price before I fill in a form."
Trust doubts: "Is this a real company? I'm not sure I'd pay here."
Giving up: "I'd probably just leave and check another site."
The mistakes that ruin most DIY usability tests
Most first tests go wrong in the same few ways:
Helping. The moment you say "It's under Services", you've lost the insight. Instead, ask "Where would you expect to find it?"
Leading questions. "Was that easy?" invites a polite yes. "How did that feel?" gets a real answer.
Testing with the wrong people. A designer friend will critique fonts. A real customer will tell you they couldn't find the delivery fee.
Defending the design. Explaining why a page is built that way turns the session into a debate. Just listen and write it down.
Acting on one comment. One person disliking a photo is an opinion. Three people missing your "Book now" button is a problem.
A quick example: the bakery that buried its order cut-off
Imagine a small Singapore bakery that sells celebration cakes online. Traffic was steady, but orders weren't growing, and the owner assumed prices were the problem.
She ran a simple test with five past customers over video call. The task: "You need a birthday cake for this Saturday. Order one."
Four out of five chose a cake, reached checkout, and only then discovered that orders needed three days' notice. Two said they'd have given up and ordered elsewhere. One said, "I'd have picked a different cake if I'd known earlier." Nobody mentioned price.
The fix took an afternoon: a clear line about lead time on every product page and a delivery date picker that greyed out unavailable days. No redesign, no new ads. Just one problem, seen five times, finally fixed.
That's the real value of usability testing. It replaces guesses with something you watched happen.
How to decide what to fix first
After your sessions, list every issue you saw. Next to each one, note how many people hit it and how badly it affected them (a small annoyance, a slowdown, or a complete stop). Then fix in this order:
Issues that stopped people completing the main goal.
Issues several people hit, even if they eventually got through.
Quick wins: unclear wording, missing prices, hidden contact details.
Make the changes, then test again with a fresh group of three to five people. Small, regular rounds beat one big project every time.
If you're seeing visitors arrive but not enquire, this pairs well with our guide on why visitors leave your website without enquiring.
Frequently asked questions
Can I usability test my own website for free?
Yes. All you need is a list of tasks, a video call tool or a quiet table, and five people who match your customers. The only real cost is your time and a small thank-you for participants. The hardest part is staying quiet and not explaining your site while they use it.
How long should a usability test session take?
For a small business website, 20 to 30 minutes per person is usually enough. That's long enough for a short introduction, three to five tasks and a few closing questions, but short enough that people stay focused. If you run five sessions back to back, you can finish in a single afternoon.
Should I test on mobile or desktop?
Test on whichever device most of your visitors use. Your website analytics will show the split. If most traffic is mobile, ask participants to use their own phone, because that's how real customers see your site. If it's split evenly, run a few sessions on each and compare where people struggle.
Is usability testing the same as a customer survey?
No. A survey asks people what they think. Usability testing watches what they actually do. Surveys are useful for reaching many people quickly, but they rarely show you the exact moment someone gets confused on your website. Testing shows you that moment, often in someone's own words. Many businesses benefit from using both.
What if I can't find people to test with?
Start with people who already know you: recent customers, newsletter subscribers or people who enquired but didn't buy. A short, friendly message offering a small thank-you usually works. Avoid friends and family where possible, because they know too much about your business and tend to be too polite.
Want a fresh pair of eyes first?
Running your own test is one of the best things you can do for your website. But if you'd like to know where to look first, we can help.
Our free UX audit reviews your homepage from a first-time customer's point of view and points out the three biggest issues likely costing you enquiries, with practical fixes.
Want to see what this looks like in practice?