I redesign websites without destroying their rankings: full URL and redirect mapping, content preservation, performance rebuild and staged cutover — because a redesign that loses your search equity is a demolition with better typography.
Redesigns kill more rankings than algorithm updates do
The pattern is consistent and entirely avoidable. New agency, new design, new URL structure, no redirect map. The site launches on a Thursday, everyone is pleased, and by the following Monday organic traffic has walked off a cliff that nobody attributes to the launch for another six weeks.
Every renamed URL without a 301 hands back everything that URL had earned — every link pointing at it, every year of accumulated signal, every ranking it held. Every merged page without a redirect orphans the links that pointed to the original. The internet does not know your new sitemap exists; it knows the old one, link by link, and it will keep sending people to addresses you deleted.
The permanence of URLs is a deal you made whether or not anyone explained it. This service is about honouring it while still getting the site you want.
What the engagement covers
- A full crawl of the existing site first. Every URL, its traffic, its inbound links, its rankings. This is the inventory of what you are protecting, and it has to exist before a single design decision is made.
- The redirect map. Every old URL to its precise successor. Never bulk-redirected to the homepage — a redirect to
/is a soft 404 with better manners, and search engines treat it that way. - Content preservation audit. The pages earning your traffic get carried across rather than “refreshed” into something shorter. Redesigns routinely delete the content that was doing the work, because it did not fit the new template.
- Performance rebuild on static-first architecture, so the new site is faster rather than merely newer.
- Staged launch and verification. Parallel checks before cutover, then a post-launch crawl and a fortnight watching Search Console coverage until the data confirms nothing broke.
The conversation worth having before you commit
Ask your prospective designer three questions. Who is producing the redirect map? What happens to the pages currently earning traffic? How will we verify nothing broke after launch?
If the answers are vague, or if any of them is “that’s a technical thing we’ll handle at the end”, you now know where the risk sits. That is useful to know whether or not you hire me — and I would genuinely rather you asked your existing agency those questions and got good answers than switched to me and got the same site.
Why the content audit matters as much as the redirects
A surprising amount of ranking damage comes not from lost URLs but from lost words. The old page had eleven hundred words answering a specific question; the new template has room for three hundred and a hero image. The URL survived, the redirect was unnecessary, and the ranking still went.
So the audit runs page by page on anything with traffic: what is it ranking for, what content is doing that work, and does the new design have room for it. Where it does not, the design accommodates the content rather than the content being trimmed to fit — which is a conversation best had during design rather than during the traffic post-mortem.
If it has already happened
Speed matters. Reconstruct the old URL inventory from your sitemap history, analytics exports, Search Console data and archive snapshots. Lay the redirects. Restore the content that was cut, where it can be recovered.
Recovery is partial and it is real — I have seen a substantial share come back from a migration that was months old by the time anyone diagnosed it. What does not come back is the traffic lost in the interval, which is why the preventive version of this page exists.
The launch sequence
Two weeks before: the crawl is complete, the redirect map is written and reviewed, and every page with meaningful traffic has a named destination. Nothing about the design blocks this — it can happen while visual work continues.
Launch day: redirects deployed with the site rather than afterwards, sitemap submitted, and a crawl run within the hour to catch anything the map missed. The first crawl almost always finds two or three stragglers; finding them on day one rather than in month three is the entire value of the step.
The fortnight after: Search Console coverage watched daily for spikes in 404s or excluded pages, rankings for your top queries compared against the pre-launch baseline, and any drift investigated while the cause is still obvious.
Month two: a formal comparison. Traffic, rankings, and indexed page count versus baseline, written down. If something is down, we find out why then rather than arguing about it a year later.
The number that decides the budget
Before quoting anything, I want to know what your organic traffic is currently worth — roughly, in revenue, per month. That figure decides how much protection is proportionate.
A site earning nothing from search can be redesigned casually and the redirect discipline is optional. A site where organic drives a serious share of enquiries deserves the full mapping exercise, because a fortnight of lost rankings costs more than the entire project. Most businesses have never calculated this, and calculating it changes the conversation from cost to insurance — and it is a calculation your analytics already contains, needing about twenty minutes, no consultant, and a willingness to look at the number honestly.
Who should do this without me
If your site has little or no organic traffic, redesign it however you like and skip every precaution on this page — there is nothing to protect, and the redirect discipline would be ceremony. Hire the designer whose work you admire and enjoy the process.
The precautions here earn their cost exactly in proportion to what search currently sends you. Knowing which situation you are in is the first decision, and it is one you can make yourself in an afternoon with your analytics open.
Related reading
/services/website-design — a new site with no traffic to protect · /services/technical-seo — the audit layer · /blog/technical-seo — redirect hygiene in detail · /blog/zero-downtime-migration — the same discipline applied to systems
Before you launch, not after
Redesigning this quarter? The before-call is cheap and the after-call is a rescue. /contact — bring the current sitemap and a launch date.
Questions I actually get
Our designer says redirects are IT's problem.
Your designer is the reason this page exists. A redesign is one project or it is two disasters — the visual work and the URL work are the same engagement, and separating them is how businesses lose years of accumulated ranking in a single weekend.
Can you rescue a redesign that already tanked?
Often, yes, if the old URL inventory can be reconstructed from sitemaps, analytics history or archive snapshots. Redirects laid late still recover a meaningful share, though speed matters because signals decay. It is a worse position than preventing it and far better than accepting it.
How much longer does doing it properly take?
The mapping adds about a week. That week is the difference between keeping your rankings and rebuilding them over the following year, which makes it the highest-return week in the entire project.
Will it look as good as an agency redesign?
It will look considered and it will load fast. If your priority is a visually ambitious brand experience, hire a design studio and bring me in for the migration layer — that division of labour works well and I will say so plainly.
What if we are changing platforms as well?
Then the redirect work matters more, not less, because platforms impose their own URL conventions and the mapping has to survive that translation. Replatforming without a map is the single most reliable way to lose organic revenue.
How do we know it worked?
A post-launch crawl, Search Console coverage watched for a fortnight, and a comparison of rankings for your top queries before and after. Verification is part of the project rather than something you hope about afterwards.