Before redesigning the page, understand the hesitation

When a page underperforms, its appearance is often the first thing questioned. Sometimes the design does make the task difficult. At other times, visitors need information, reassurance or a different offer. Conversion research helps identify which problem the redesign is supposed to solve.

Without that diagnosis, teams can spend substantial effort replacing a page and still preserve its central weakness. A new layout cannot answer a question that the content continues to avoid.

Observe the task, not just the page

Choose the action a suitable visitor should be able to complete. It might be comparing a service, understanding a product or submitting an enquiry. Review the journey that leads to it, including the ad or search result that sets expectations.

Ask people to attempt the task and explain what they understand. Avoid leading them towards the answer the team wants to hear. A hesitation around price may actually reflect uncertainty about scope; a reluctance to submit a form may come from not knowing what follows.

Combine different kinds of evidence

Use available analytics to identify where investigation is needed, then combine it with customer questions, support feedback and observed behaviour. No single source tells the whole story. A high exit rate does not explain why someone left, and one interview does not establish how common a problem is.

AI can organise approved notes and surface recurring language. Keep the original observations available and review whether the synthesis has flattened important differences. Small samples are useful for discovering problems, but should not be presented as precise population estimates.

Translate a finding into a hypothesis

Write down the observed issue, the proposed change and the expected mechanism. For example: visitors cannot distinguish two packages, so a clearer comparison of scope may help them select a suitable starting point. That is a more testable idea than making the page feel more premium.

Prioritise by likely importance, confidence in the evidence and implementation effort. Include technical defects and accessibility barriers, which may warrant fixing directly rather than waiting for a marketing experiment.

Check whether the change helped

Choose an evaluation method suited to the traffic and question. A controlled experiment may be appropriate when conditions support one; a structured usability review may be more useful for a low-volume service page. Avoid claiming certainty that the evidence cannot support.

Keep a record of what changed and what was learned. Conversion improvement becomes cumulative when the team understands the reasons behind its decisions. The next redesign can then build on customer knowledge instead of starting again with a new set of visual preferences.