Stop redesigning. Start diagnosing.

A website stops performing. Enquiries drop, or they never really started. Somebody says the site looks tired. Within a month there is a proposal for a full redesign, a five-figure number, and a twelve-week timeline.

Nobody in that sequence established what was wrong.

The redesign is a diagnosis by other means

Rebuilding a site changes every variable at once — structure, copy, imagery, hosting, template, forms, tracking. If enquiries go up afterwards, you cannot say why. If they go down, you cannot say why. You have spent the budget and learned nothing you can reuse.

And the change most likely to have mattered is usually one you could have made on the old site in an afternoon. We see this repeatedly: the rebuild fixed the 4 MB hero image and moved the phone number above the fold. Everything else was decoration.

Tired is not a diagnosis

"It looks dated" is a real observation and a poor brief, because it does not distinguish between the site looking unfashionable to you, who see it daily, and the site failing customers, who see it once.

The failures we actually find, over and over:

• The pages people arrive on are not the pages anyone designed. Traffic lands on an old service page from 2019 that the redesign brief does not mention.
• The form works on desktop and silently fails on iOS. Nobody tested it on a phone; enquiries have been dropping for months and it reads as a marketing problem.
• The site is slow on mobile only, so everyone internally thinks it is fine.
• The thing the business actually sells is on page four.

None of those require a redesign. Two of them survive one.

What to do instead, in order

Find out what is broken. Performance, structure, accessibility, and whether the forms actually deliver mail. A day of work, or an automated audit in under a minute.

Find out what people do. Which pages they land on, where they leave, what they search for on your own site. Your internal search log is the cheapest customer research available and almost nobody reads it.

Fix the three things that are demonstrably wrong. One at a time, so you learn which one mattered.

Then decide about the redesign — with evidence, and often for reasons that have nothing to do with performance. Sometimes the brand genuinely has moved on, or the platform is a dead end, or you cannot edit your own pages. Those are good reasons to rebuild. "Enquiries are down" is not, until you know why.

For agencies, this is the better sale

Leading with a diagnosis rather than a rebuild feels like leaving money on the table. In practice it converts better, because you arrive with findings instead of opinions, and the first conversation is about their business rather than your portfolio.

It also protects you. A rebuild sold against a vague complaint is a rebuild that will be judged against a vague standard, and you will be the one holding it when the numbers do not move.

Diagnose first. Then rebuild, if the diagnosis says so.