Geeks Digital Client identity withheld. Figures compare post-recovery performance to the pre-migration baseline year.
The situation
An established store replatformed over a weekend. Nobody mapped the URLs. Monday morning, thousands of product and category addresses returned 404s, and within six weeks organic revenue was down by more than half. The previous developer’s advice was to wait it out. Waiting was making it worse: Google was steadily forgetting pages that had taken six years to earn their positions.
What we found
Forensics on the old crawl data, analytics history and the new platform’s structure produced the damage map. Roughly 3,800 URLs with rankings or backlinks now died at a 404. The new platform had also introduced its own problems while nobody was looking: faceted navigation generating tens of thousands of crawlable filter URLs, duplicate product pages under multiple category paths, and a template that shipped its content client-side. The store had two diseases: acute equity loss and a chronic architecture problem the migration had smuggled in.
The strategy
Triage before growth. First, reconnect the old equity through redirects prioritized by value, because every week of delay let more of it evaporate. Second, stop the new platform from digging fresh holes. Only then invest in growth, on foundations that could hold it.
The work
Phase 1: the redirect rescue (weeks 1-3)
Old URLs were tiered by traffic, rankings and backlinks. The top tier was mapped by hand to exact equivalents; the middle tier through pattern rules with spot verification; the tail to closest category parents. Legacy sitemaps were temporarily resubmitted to push recrawling of the redirects. Within days, Search Console showed the 404 curve bending.
Phase 2: stop the bleeding the platform caused (weeks 3-8)
Facet crawling was brought under control with parameter rules and selective noindexing. Canonical logic was fixed so each product lived at one address. The template’s critical content moved server-side, making the catalog readable to crawlers and, incidentally, to the AI systems that would matter later.
Phase 3: prune and rebuild (weeks 6-16)
The catalog carried hundreds of dead products and thin duplicate pages the migration had exposed. Discontinued items were redirected to living alternatives, thin variants consolidated, and the surviving category pages rebuilt with genuine buying guidance and internal links that concentrated authority where margin lived.
The results
Rankings recovered on a curve that tracked the redirect waves: 96% of previously held top-ten positions were back within nineteen weeks. The chronic fixes then did what the old site never could: twelve months after the engagement began, organic revenue stood 64% above the pre-migration baseline year, on a catalog that indexed cleanly and loaded fast. The disaster ended up funding the rebuild the store had always needed.
What made the difference
Speed and sequencing. Migration equity decays on a clock; the value-tiered redirect rescue in week one saved what a “wait and see” quarter would have buried. And treating the migration as an opportunity to fix the architecture, not just restore it, is why the recovery became growth.
Explore the services behind this engagement: Technical SEO, Ecommerce SEO and SEO Audits.
