Redesign, same addresses
New look, same URLs and content.
Best for: Lowest risk. Watch for content, headings and internal links that get dropped in the redesign.
A site migration is any big change to a website's addresses, platform or structure: a redesign, a move to Shopify, a new domain. Done badly, it's the fastest way to lose years of Google rankings in a week. Done properly, the traffic carries over. The core of it is a redirect map that sends every old address to its closest new one, checked before launch, tested on the day, and watched for weeks after.
From my audits
Platform moves, category restructures and full rebuilds tend to go wrong in the same places. These are the problems that come up most.
The new site is designed, built and nearly ready before anyone checks what the old site ranked for. By then, the URL structure is set.
Main pages get redirected. Old blog posts, PDFs, image URLs and product variants with links pointing at them don't.
Category pages lose the text that made them rank because the new design 'looks cleaner'. Rankings follow the text out.
The test site is hidden from Google, which is right. Then the block goes live with the new site.
Analytics and conversion tags don't make it to the new site, so nobody can tell whether launch helped or hurt.
What Google says about moves
Google publishes how it handles a site move. Most migrations that lose traffic break one of these.
Google says it's important to map your old site's URLs to the new site's URLs, each to its closest match, before setting up redirects.
What I do: Build the map from a crawl of the old site, Search Console and analytics, so nothing that earns traffic or links is missed.
Google says not to redirect many old pages to one irrelevant page like the home page, because it may be treated as a soft 404: a page Google treats as missing.
What I do: Send each page somewhere that answers the same question. If there's truly nowhere, let it 404.
A permanent (301 or 308) redirect tells Google the new address should be the one shown in search. A temporary one keeps the old address in play.
What I do: Use 301s for a migration, and keep them for at least a year, ideally longer.
Google's crawlers follow up to 10 redirect hops, and every hop slows people down. Old redirects pointing at new redirects build chains over the years.
What I do: Point old redirects straight at the final address, not through the last migration's redirects.
Google says rankings may fluctuate while it recrawls the site, and a medium-sized site can take a few weeks or more to settle.
What I do: Plan the launch away from your busiest season, and watch Search Console daily for the first fortnight.
Checked against Google Search Central on the date below. Every rule is linked, so you can show your developer.
Redirect map check
The redirect map is where most migrations win or lose. Paste yours: one old address and its new address per line, separated by a space, comma or tab. It flags the mistakes before they go live.
Problems found
0
Paste a map, or load the example.
Sending several old pages to one page is fine when their content really was combined into it. Sending them to the home page because there's nowhere better is the problem.
The human side
Rankings aren't the only thing that dips after a relaunch. Returning customers do too.
01Launch week
Google's own researchers describe change aversion as the negative short-term reaction to changes in a product, and expect satisfaction to recover once people get used to the new version.
A dip in enquiries in launch week isn't always an SEO problem. Regular customers may just be lost.
What I do: Tell customers before the change, keep the things they use most easy to find, and judge the new site after the settling-in period, not on day three.
Linked so you can read it. It's a product research finding, and it applies to websites as much as apps.
Kinds of migration
The risk depends on what's changing. The more that changes at once, the harder it is to tell what caused a drop.
New look, same URLs and content.
Best for: Lowest risk. Watch for content, headings and internal links that get dropped in the redesign.
For example WordPress to Shopify, or an old system to a new one. Addresses usually change.
Best for: Medium to high risk. Product and category URLs change pattern, so the map matters most.
Categories merged or split, URLs renamed, sections moved.
Best for: Medium risk. Done well, it can lift rankings. Done badly, it splits pages that used to rank.
A new domain name, or merging two sites into one. Needs Google's Change of Address tool as well as redirects.
Best for: Highest risk. Everything moves at once, so every signal has to transfer.
How it works
What the old site ranks for, which pages earn traffic, links and enquiries, so nothing valuable is lost.
The new structure based on real searches, and a redirect map for every address that matters.
Redirects, indexing, robots rules, canonicals, sitemaps, tracking and Search Console, tested on the day.
Daily checks for two weeks and weekly after, fixing anything that slips before it costs rankings.
What you get
Signals this works on
Out of the 8 signals every page and ad sends.
You don't have to. Google says rankings may fluctuate for a while during a move. A good redirect map and keeping the content that ranked usually means a short wobble, not a lasting loss.
Google recommends at least a year, and longer if you can. In practice, keep them as long as the site exists, especially for pages other sites link to.
Before the new structure is designed. Changing URLs, categories and page content is what moves rankings, and those are decided early.
Yes. Use the Change of Address tool in Search Console as well as redirects. It isn't needed for moves within the same domain, like HTTP to HTTPS.
For a medium-sized site, Google says it can take a few weeks or more to show the new addresses. Large sites take longer.
That the old addresses redirect to the right new ones, the new site isn't blocked from Google, canonicals and sitemaps point at the new addresses, and analytics and conversion tracking are firing.
Takes a minute. Audits come back within 1 to 2 business days. For anything else, I'll get back to you as soon as possible.
What happens next
Keep going