Website redesign and modernisation

What this actually solves

The site was built years ago by someone who is no longer reachable. On a phone it is unusable — text too small, a menu that does not open, a form that runs off the edge. In Google it has been drifting down without any single dramatic drop. And underneath all of that, the reason people rarely say out loud in the first email: the owner has stopped sending the link. When someone asks for the website, there is a small hesitation before they give it. That hesitation is expensive in a way no analytics dashboard records.

There is usually a second trigger too, and it is almost always a competitor. Somebody in the same trade relaunched, it looks current, and the comparison is now unavoidable to anyone choosing between you.

A redesign is a different job from building a new site, and the difference is the whole point. On a new site there is nothing to lose. Here there is: URLs that other websites already link to, pages that rank for things you did not know you ranked for, and a history Google has spent years building an opinion about. The craft in a redesign is not the new design — it is carrying the old value across intact. Sites get rebuilt beautifully and lose half their traffic every week, and the cause is nearly never the design.

What we do, and in what order

1. Crawl the old site completely, before anything is touched. Every URL, exported and kept, using Screaming Frog — free up to 500 URLs, and anything larger needs the paid version or a crawl in two passes. This is ten minutes of work and it is the single step that saves the project, because once the old site is replaced, the list of what used to exist is gone and cannot be reconstructed from memory. Half a day.
2. Find out what each page is actually worth. Search Console and Analytics, page by page, for traffic, impressions and inbound links. This is where redesigns get their surprises — it is common to find most organic traffic landing on two or three old blog posts nobody has thought about in years, written for a question customers are still asking. Deleting those because they look dated is how a redesign becomes a loss. The output is a keep, merge, rewrite or retire decision against every URL, with a reason. One to two days.
3. Design the new structure, then the pages. A new sitemap first — what the site should contain now, not what it contained in the year it was built — then wireframes for the page types, then visual design on top of them. Mobile is designed first here rather than adapted afterwards, because the traffic already is. One to two weeks, including your review rounds.
4. Build, and write the redirect map alongside it. Every old URL gets a destination written down next to it. Straight to the final page, never through an intermediate hop. Query strings and trailing-slash variants are handled explicitly. The homepage is a last resort for orphans, never the default answer, because sending an old article to the homepage tells Google the article is gone and tells the visitor they were wrong to click. One to two weeks.
5. Launch, then watch for two weeks. The redirects are tested by re-crawling the old URL list against the live site and confirming every one lands where it should. Then Search Console's coverage and 404 reports are checked repeatedly over the following fortnight, because links you never knew about surface only when someone follows one. Anything that appears gets a redirect added. Half a day at launch, then periodic checks across two weeks.

What moves the price

The range is $1,000–$6,000, and three things decide where a project sits inside it.

Page count, more than anything else. Forty pages and four hundred pages are not the same project with a different number on it. The design work scales gently; the per-page decisions, the redirect mapping and the post-launch verification scale directly, and those are the parts that cannot be skipped.

Whether the words are kept or rewritten. Migrating existing copy into a new structure is production work. Rewriting it is a different service with its own timeline, and mixing the two silently is how redesigns run over. It is worth deciding this per page rather than for the whole site.

How much traffic the current site has to protect. A site with real organic traffic demands far more care than one nobody visits — more analysis before the cut, more careful redirect work, and more verification after. That caution is most of the difference between the bottom and the top of the range, and it is the correct place to spend money if the current site earns anything at all.

What usually goes wrong

The old site is replaced before it was ever crawled. This is the one that ends relationships. The new site goes live, everyone is pleased, and three weeks later organic traffic has halved. You ask what happened, and there is no answer available, because the list of URLs that used to exist died with the old site and nobody can now say what should have redirected where. The fix costs ten minutes and has to happen on day one.

Pages are judged by how they look rather than what they earn. An old article with dated formatting and no images gets cut in the tidy-up. It was the page bringing in a third of the search traffic. Nothing in the new design is at fault; the decision was made on taste without opening Search Console first, and taste has no opinion about inbound links.

Redirect chains and a homepage catch-all. Old page to interim page to final page passes through two hops, and each hop is a chance for something to be dropped and a delay for the visitor. Worse is the blanket rule that sends every unmatched old URL to the homepage — it looks like thoroughness, it satisfies the checklist, and it throws away the specific relevance that made those links worth having.