Website Usability Testing: Watch Before You Redesign

It is surprisingly difficult to see your own website as a newcomer. You know what the service names mean and where the contact link lives. Website usability testing lets you find out what happens when someone arrives without that knowledge, before you spend money fixing what the team merely dislikes.

The revealing moment is often a small hesitation. Someone reads the same label twice, goes back a page or asks whether they are in the right place. Resist the urge to explain. That uncertainty is useful evidence about what the website needs to do better.

Give website usability testing a narrow question

Choose one decision you need evidence for. Perhaps people cannot identify the right service, or they miss the route to support. A narrow task makes the session easier to run and the findings easier to apply. You can explore another journey in a later round.

Imagine a commercial catering business with several event services. A realistic task might ask someone to find a suitable option for a workplace function and explain how they would begin arranging it. Avoid naming the menu item you expect them to click.

Define what completion looks like without prescribing a route. The purpose is to learn how the participant interprets the site, including routes the internal team may not expect.

Recruit people who resemble the intended audience

Find people who could plausibly need the service and do not already know the site inside out. Helpful colleagues often fill in gaps without noticing. In user experience research, that familiarity can hide the very confusion you are trying to uncover.

A small round can reveal useful issues, although it does not measure how common those issues are across the entire audience. Treat it as qualitative user experience research rather than a statistically representative survey.

Explain the session and any recording or information collection clearly. This article is general information only, not legal or privacy advice. Obligations depend on the organisation, industry, jurisdiction, technology and circumstances. Review current regulator guidance or obtain independent professional advice where appropriate.

Observe without taking over

Let the silence last a little longer than feels comfortable. Ask what the person is looking for, then listen. If you explain where to click, you learn that your instructions work; you still do not know whether the website can carry the task on its own.

Record observations separately from interpretations:

• Observation: The participant opened three service pages before choosing one.
• Interpretation: The service labels may not distinguish the options clearly.
• Possible improvement: Add a short comparison near the service entry point.

That separation helps prevent one uncertain interaction from becoming an overconfident design conclusion.

Prioritise issues by their effect on the task

Separate a preference from an obstacle. Disliking a colour is different from being unable to identify the appropriate service. If the issue is unclear scope, better service page copy may address it. Rank the findings by their effect on the task before writing a redesign wishlist.

For the catering business, several participants might understand the service descriptions but miss the distinction between delivery-only and staffed events. A clearer comparison could be tested before rebuilding the whole website.

Include content changes among the possible responses. Problems that appear visual can sometimes be resolved through clearer labels or better information order.

Test the improvement, not just the original problem

Return to the same task after making a change, ideally with fresh participants. The aim is to see whether the explanation or route has improved. Someone who remembers the previous session may succeed because they now know the answer, even if the site remains confusing.

Keep the task comparable and record what changed. This creates a useful basis for choosing the next improvement rather than relying on the team's preference for the new layout.

Do we need a complete redesign to act on findings?

A complete redesign is not the inevitable result. Navigation labels, explanations and next steps can often be improved within the existing site. If the findings point to larger structural work, carry the evidence into your website brief so the rebuild has a specific problem to solve.

Use evidence to narrow the work

For a first session, write one realistic task and invite someone suitable to attempt it without coaching. Record what they do, where they hesitate and what they expect. That short observation can give the next website discussion something more useful than competing opinions.

Use the findings to inform the wider experience, with further context in why a clear UX strategy matters. Explore our website services when the research points to changes that need design and implementation support.