A website migration can look like a success on launch day, then quietly damage enquiries over the following weeks. This SEO recovery after migration example shows what that looks like in practice: traffic fell, high-value pages slipped out of Google, and the real issue was not one dramatic technical failure. It was a series of small gaps that added up.
The point is not that every migration follows this pattern. A small brochure site and a large eCommerce platform carry different risks. But the recovery process is broadly the same: establish what changed, protect the pages that bring commercial value, remove the barriers preventing Google from understanding the new site, then measure recovery against leads rather than vanity metrics.
The migration problem: traffic was down, but not everywhere
In this example, an established UK service business replaced an ageing website with a faster, better-designed site. The new site was easier to use, looked more credible and gave the sales team stronger landing pages. On paper, it was an improvement.
Three weeks after launch, however, organic sessions were down by 34%. More concerningly, form enquiries from organic search had dropped by nearly half. Brand searches were holding up, but non-brand terms – the searches made by people looking for the service rather than the company name – had lost visibility.
This distinction mattered. Looking only at total traffic might have encouraged everyone to wait it out. Looking at the pages and keywords that had previously generated enquiries showed a more urgent problem. Several former top-performing service and location pages were no longer ranking as they had before.
The first mistake businesses make here is assuming a new website needs time to settle. Sometimes it does. Google has to crawl, process and reassess changed pages. But a significant fall in commercially useful traffic deserves an investigation, not blind patience.
SEO recovery after migration example: finding the real causes
The recovery began with a comparison of the old and new sites. We reviewed historical organic landing-page data, conversion data, crawl reports and indexed URLs. That created a clear view of which pages had driven value before the migration and what had happened to them afterwards.
Four issues were responsible for most of the loss.
Redirects existed, but they were not precise enough
The site had redirects in place, which was a good start. The problem was that many old pages were sent to broad category or service pages rather than the closest relevant replacement. A location-specific service page, for example, was redirected to a general service page.
From a user perspective, that was not always terrible. From a search perspective, it weakened relevance and discarded useful page-level signals built up over time. A redirect should tell Google, as clearly as possible, where the old page now lives. Sending dozens of distinct pages to one destination creates uncertainty and can look like a poor substitute for the original content.
The fix was to map each valuable old URL to its closest new equivalent. Where no proper equivalent existed, we rebuilt the missing page rather than forcing a weak redirect.
Key content had been shortened too far
The redesign had made several service pages cleaner, but also thinner. Useful supporting copy, specific examples, FAQs and location context had been removed in favour of a more minimal layout.
Minimal design is not automatically bad for SEO. Neither is longer copy automatically better. The issue was whether each page still answered the searcher’s question and showed why the business was a credible choice. In this case, the new pages had lost much of the detail that made them relevant to non-brand searches.
We restored the strongest information from the previous site, then improved it rather than copying it blindly. That included clearer service scope, practical buying considerations, local relevance where it genuinely applied, and stronger calls to action. The goal was not to add words for the sake of it. It was to make the page useful to someone deciding whether to enquire.
Internal links no longer supported priority pages
The old site linked prominently to important services from related pages, navigation areas and resource content. The new design reduced many of these links or replaced descriptive anchor text with generic buttons.
Internal links help visitors move through a site, but they also help search engines understand which pages matter and how subjects relate. The lost links had made it harder for Google to discover and assess several priority pages.
The revised internal linking plan added natural routes from relevant services, case studies and supporting content into the pages that generated leads. Navigation remained simple. The work focused on useful context, not stuffing every page with repetitive links.
Indexing signals were inconsistent
A handful of migrated pages had incorrect canonical tags, pointing to old URLs or to broader pages. Some low-value parameter URLs were being crawled unnecessarily, while a small number of important pages had been omitted from the XML sitemap.
None of these issues alone explained the full drop. Together, they made the migration harder for Google to interpret and slowed recovery. Canonicals were corrected, priority pages were included in the sitemap, and crawl waste was reduced. We also checked that no staging rules, noindex tags or blocked resources had made their way onto the live site.
What happened after the fixes
The work was prioritised by commercial impact. First came redirects and indexing issues, because there was little point improving content if Google could not reliably access or associate the right pages. Next came restoring missing service and location content. Internal links and conversion improvements followed.
Within two weeks, Google began recrawling the corrected redirects and pages. Rankings did not all return at once. Some terms recovered quickly because a clear like-for-like destination now existed. Others took longer, particularly where the new page needed to regain relevance after substantial content changes.
By week eight, organic sessions were 18% above the level before the migration. More importantly, organic enquiry volume had increased by 27%. The redesigned pages converted better than the old ones once their visibility returned.
That is the commercial lesson. Traffic recovery is useful, but it is not the finish line. If a website is rebuilt to support growth, the measure should be qualified leads, sales opportunities and revenue where tracking allows. A migration can be technically tidy yet still underperform if the new site makes it harder for prospects to take the next step.
A practical recovery plan for your own site move
If rankings or leads have dropped after a migration, start by comparing the last stable period before launch with the weeks after it. Segment branded and non-branded traffic, then identify the landing pages that lost the most clicks, impressions and conversions. This prevents the team wasting time on pages that were never commercially important.
Check that every meaningful old URL returns a single 301 redirect to the most relevant live page. Avoid chains, where one redirect leads to another, and avoid sending everything to the homepage. Then crawl the live site to find broken links, incorrect canonicals, accidental noindex tags, poor internal linking and pages missing from the sitemap.
After the technical checks, review whether priority pages still deserve to rank. Has useful content disappeared? Have unique service, product or location pages been merged into generic pages? Is the main offer clear, credible and easy to act on? These are SEO questions, but they are also sales questions.
It also helps to separate migration recovery from wider SEO activity. Do not redesign every underperforming page while trying to diagnose a traffic drop. Stabilise the migration first. Once the foundation is sound, ongoing organic SEO and AI SEO work can build visibility across the topics and questions your buyers are actively researching.
How to avoid needing a recovery plan
The best time to protect organic performance is before development starts. Build a URL inventory, identify pages with traffic, rankings, backlinks and conversions, and agree their future purpose before anyone removes or combining content. Redirect mapping should be treated as part of the build, not an admin task on launch day.
A pre-launch crawl of the staging site can catch many costly issues, but it is not enough on its own. The launch needs checks too: redirects, robots rules, indexability, canonical tags, sitemap coverage, analytics, conversion tracking and the live user journey. A polished new website that cannot be found or cannot generate an enquiry is not finished.
A migration does not have to cost a business months of visibility. But recovery is faster when decisions are based on evidence, the strongest pages are protected, and the team focuses on the search traffic that actually contributes to growth. If your site has recently moved and the numbers do not look right, start with the pages that used to bring the best enquiries. They will usually tell you where the real problem is.