Website redesignWebsite redesign

Why most redesigns lose traffic

The design is rarely the problem. A redesign loses traffic because the URLs changed and nobody wrote the redirects, so every link Google had — and every link anybody else had — now points at a page that no longer exists. Google drops those pages from the index over the following weeks, and the drop looks like a mysterious algorithm hit rather than the self-inflicted wound it is.

The second cause is quieter and harder to reverse: content that was cut for looking dated. A long, slightly ugly page that answers a question thoroughly is often the page earning the traffic. Replacing it with three lines and a photograph is a design improvement and a commercial loss at the same time.

• URLs changed with no redirect map — the single most common cause
• Long pages cut down to look cleaner, taking their rankings with them
• Title tags and headings replaced with brand slogans instead of what the page is about
• Internal links dropped, so the important pages are suddenly three clicks deep
• Analytics and Search Console not reconnected, so the drop is invisible for a month

None of that is difficult. It is simply work that produces nothing visible, on a project everybody is judging by how it looks.

What a website redesign costs in the UK

Here, a focused redesign starts at £500 plus £50 a month for hosting and care. That is a genuine starting point rather than a headline: a small business site with a handful of pages, rebuilt properly, with the redirects done. Large e-commerce rebuilds and anything with bespoke functionality are priced on application, from the hours the work takes, because a 4,000-product catalogue and a six-page brochure are not the same job and pretending otherwise produces a quote that is wrong in one direction or the other.

Nothing here is built from a bought theme. That matters to the price in both directions — it costs more hours than installing a template, and it costs fewer hours than the eighteen months of fighting a theme’s assumptions that usually follows one.

The redirect map, which is the actual job

Before anything is designed, the existing site is crawled and every URL is listed with the traffic and links it currently earns. That list becomes a map: old URL, new URL, and a note where there is no equivalent. Every row gets a 301 redirect. Rows with no equivalent get a decision made by a person rather than a catch-all to the homepage, which Google treats as a soft 404 and which helps nobody.

This is also where an inherited site’s history stops being somebody else’s problem. Sites that have been redesigned twice already often carry chains — page A redirects to B which redirects to C — and each hop loses a little. They get flattened to a single redirect at the same time.

Beds.ie is the evidence for this working at scale rather than in theory. The rebuild kept and reorganised a large catalogue, and organic traffic went from 672 clicks in April 2026 to 7,781 in August, with impressions from 38,608 to 282,069 — Search Console figures, full calendar months, read on 4 September 2026. A redesign that loses rankings does not produce that shape.

When you should not redesign

If the site converts, ranks and loads quickly, and the only complaint is that it looks a bit 2019, a redesign is an expensive way to buy a feeling. The same money spent on the checkout, the enquiry form or the three pages that already get traffic will usually return more.

The honest test is whether you can name what the redesign is supposed to change. “More enquiries” is a target; “fresher” is not. If the answer is the second one, it is worth a conversation before it is worth a quote, and sometimes the outcome of that conversation is that you keep the site.