Here is the website redesign story nobody tells you at the proposal stage: a business runs a site for four years, quietly accumulating rankings, reviews of pages in Google’s memory, and a steady stream of organic enquiries.

Then it launches a beautiful new website, and the enquiries stop. Not dip: stop. Because somewhere between the old site and the new one, the URLs changed, nothing redirected, and four years of accumulated search equity, the asset that takes years to build, was deleted in an afternoon and celebrated with a launch post.

A redesign is a heart transplant, not a paint job. Done properly, the patient comes out stronger. Done casually, the operation succeeds and the patient dies. This guide is the difference: when a redesign is actually warranted, and the migration discipline that protects what your old site earned.

First: do you actually need a redesign?

Redesigns get commissioned for the wrong reason more than any project in web work: the owner is bored of the site. Your customers see it for minutes; you see it daily, your boredom is not data.

The legitimate triggers are functional: the site fails on mobile or loads slowly on Nigerian data (the trust and speed failures that tax every marketing naira); it no longer reflects what the business actually sells; its structure can’t support what’s next (commerce, content, booking); or measurable decay, conversion falling while traffic holds.

And the honest alternative deserves a fair hearing first: most “redesign problems” are retouch problems, speed optimisation, updated copy and photography, a rebuilt homepage and key landing pages, delivered at a fraction of the cost and none of the migration risk. Rebuild when the foundation is wrong, not when the paint is tired.

The asset you’re carrying across (and how it gets dropped)

Your existing site owns things the new design doesn’t automatically inherit: every URL Google has indexed and ranked, every backlink and press mention pointing at specific pages, every page’s accumulated history, plus the practical wiring: analytics, tracking pixels, form integrations, the measurement your decisions run on.

The classic disasters, in descending order of carnage: changed URLs with no redirects (the old addresses die, their rankings and backlinks die with them, this single omission is behind most post-redesign traffic collapses); thinned content (the designer trims “all that text” for aesthetics, deleting exactly what the pages ranked for); a prettier but heavier site (new animations, uncompressed imagery, slower than what it replaced, on the connections that punish it most); and the wiring left behind (analytics and conversion tracking missing at launch, so the damage isn’t even visible until the invoices are). None of these are exotic. All of them are absences of a checklist.

The migration commandment that prevents most of it: every URL on the old site either keeps its address or 301-redirects to its single closest equivalent on the new one, mapped page-by-page in a spreadsheet, before launch, no exceptions. Not everything to the homepage (that’s a soft version of deleting them), and not “the important pages”, the spreadsheet is the job. If your redesign proposal doesn’t contain the phrase “redirect map,” you’re not buying a migration; you’re buying a demolition with nice renders.

The protective sequence

1. Benchmark before anything. Export current rankings and top organic pages (Search Console), traffic and conversion baselines, page speed scores, and a full URL inventory. You can’t protect, or prove damage to, what you never measured.
2. Build on staging, launch in one cut. The new site is assembled and tested privately (your quote checklist should have demanded staging anyway), with the redirect map loaded and every form, pixel, and analytics tag verified on staging, then switched over in one controlled move, not migrated live piece by piece.
3. Carry the content, improve it, never thin it. Ranked pages keep their substance; improvements add clarity and proof rather than subtracting words the rankings were built on. Redesign the clothes, keep the muscles.
4. Launch-day audit: crawl the new site for broken links and redirect gaps, confirm analytics and conversion events are firing, resubmit the sitemap, and spot-check that top old URLs land where the spreadsheet says.
5. Watch the graph for eight weeks. A modest, temporary wobble as Google re-crawls a changed site is normal and recovers; a cliff that doesn’t is a broken migration, and week two is when it’s cheaply fixable, not month four when the quarter’s numbers ask why. The benchmark from step one is what makes the difference visible.

A website redesign done well is genuinely valuable, faster, clearer, more trustworthy, built for what the business is becoming. But the value lives entirely inside a discipline the pretty mockups never show: redesign only for functional reasons, retouch when retouching suffices, and treat migration as the actual project, benchmark first, map every URL to its heir, build on staging, carry the content whole, verify the wiring, and watch the graph.

Your old site spent years earning things Google doesn’t hand back. The new one should inherit that estate, not burn it for the insurance. Make the redirect map the star of the proposal, and the launch post will still be true in month three.

Planning a redesign, or recovering from one? The free marketing plan includes the audit either way: what your current site has earned and must keep, whether you need a rebuild or a retouch, and the migration checklist priced against what a broken one costs. If you’re spending ₦1M+ a month on marketing, it’s yours at no cost.

Take the 2-minute diagnostic →

This guide reflects standard search-migration practice and the failure patterns documented across published SEO migration research and our own project work, current as of mid-2026; the severity of post-redesign traffic loss varies with site size, how much equity existed, and how much of the checklist was skipped. This article is general information, not a guarantee of results.