“We need a better website” is a reasonable starting point, but it is not a design brief. Better might mean fewer repeated questions, more suitable enquiries, easier reservations, or a clearer way to present specialist work. Those goals lead to different pages.

Before discussing visual references, choose the problem the site should help solve. Write it in terms of an action someone can take or information they can find. This gives the project a direction that survives changes in layout, imagery, and personal taste.

Choose a meaningful next step

A service business may need visitors to describe their project. Asking for an immediate purchase would make little sense if the work requires an estimate. A restaurant needs to make its menu and booking route obvious. A portfolio may need to explain the thinking behind a project before inviting a conversation.

Choose one main next step for each page. Other links can remain available, but they should not all compete for equal attention. A visitor who is still comparing options needs evidence and context before a prominent button becomes useful.

Answer the questions that slow decisions

Useful business content is often practical. Which areas do you serve? What is included? Who will do the work? What happens after an enquiry? How long should someone expect to wait for a reply?

These details can matter more than another broad statement about quality. Where you show past work, explain the starting problem, the decisions made, and the delivered result. Separate measured outcomes from observations about the design. A well-organised interface is a valid outcome; an invented increase in revenue is not.

Define success before collecting numbers

Pick a small set of signals that relate to the goal. Completed booking requests may be useful for a restaurant. For a studio, the quality of project enquiries can be more informative than the total number of contact-button clicks.

Decide what counts and keep the definition stable. A button click is not a completed enquiry, and a page view is not evidence that the visitor understood the offer. If the final step happens in a different tool, be clear about what your website data can and cannot show.

Treat the launch as a starting point

After launch, check the actual path from page to action. Read the enquiries, test the forms, and note recurring questions. Low traffic rarely provides enough evidence for sweeping conclusions, but it can still reveal a broken link or a confusing instruction.

Choose the next edit based on something observed. If people repeatedly ask whether a service covers their area, improve that answer before redesigning the entire page. A useful website develops through decisions that can be explained, checked, and revised as the business changes.