Migration SEO.
The riskiest work in SEO,
done right.
Replatforms, redesigns, rebrands, domain changes. The process that protects organic rankings, pipeline value, and a decade of compound investment through a change that most agencies treat as a developer's problem. Pre-launch baseline, URL mapping, 301 strategy, schema preservation, eight-week post-launch monitoring. From £3,500.
The four migration types,
and how each one breaks rankings.
Migration SEO sounds technical, but the failure modes are almost always organisational. A development team focused on shipping a new build sees a successful launch as the deployment going live without errors. The SEO discipline measures success differently: no ranking position lost, no organic page disappearing from the index, no canonical tag pointing at the wrong URL, and no schema block silently dropped. Those four lenses rarely overlap with what a developer is checking on launch day.
Replatforms — the URL trap
Replatforming means moving from one CMS or framework to another. Shopify to Saleor, WordPress to Next.js, Magento to BigCommerce, Wix to WordPress. The classic failure mode is URL pattern drift. Shopify uses/products/[handle]; a Next.js build might use /shop/[slug]. If that change goes live without a 1-to-1 redirect map, every product URL Google has indexed for the last three years becomes a 404, and the link equity built on those URLs is burnt. We see 30 to 60% organic traffic loss in week one when this is missed, and full recovery typically takes four to nine months.
Redesigns — the schema and template trap
A redesign typically keeps URLs intact but rebuilds the templating. That sounds safer for SEO, and it usually is — but the failure mode shifts to schema and template-level on-page signals. The new design ships without the JSON-LD blocks the old templates emitted, or the H1 hierarchy changes (one developer's preference for visual H2-as-headline silently demotes the page's topical signal), or breadcrumb structure disappears. None of these break any tests, but they erode rankings over four to twelve weeks as Google re-evaluates the templates.
Rebrands — the trust trap
A rebrand involves a domain change. Old domain, new domain, full 301 from one to the other, GSC change-of-address request, and (this is the part most rebrands miss) a deliberate plan to maintain the old domain's SSL certificate, hosting, and redirect rules for at least eighteen months. Letting the old domain lapse three months after a rebrand because “everyone uses the new URL now” is the most expensive single decision a marketing team can make. The redirect chain dies, every backlink Google has ever crawled to the old domain returns a connection error, and the equity transfer collapses.
Domain changes — the hreflang trap
A domain change can also mean splitting one site into multiple regional properties: a single .com becoming a .com plus a .co.uk plus a .ca. Hreflang reciprocity is non-negotiable here. If the .com page links to the .co.uk variant via hreflang en-GB but the .co.uk does not link back to the .com via hreflang en-US, Google does not honour the relationship, and you get duplicate-content cannibalisation between regional variants. We see this fail on roughly half of all multi-region launches we review post-fact.
Six phases. Fourteen weeks.
Most rankings preserved.
A migration SEO engagement runs on a six-phase calendar. The pre-launch window is six weeks, launch is one week, post-launch monitoring runs eight weeks, total fifteen weeks. We can compress to twelve in a hurry; we have done it in nine, but the loss-prevention probability drops materially below that floor.
Phase 1 — Pre-launch baseline (week 1)
Full Screaming Frog crawl of the existing site, Search Console sixteen-month export, Ahrefs backlink profile, rank-tracker snapshot of every commercially relevant query, schema inventory of every JSON-LD block currently emitted. This is the artefact everything else is measured against. Without it, recovery becomes guesswork.
Phase 2 — URL mapping (weeks 2–3)
Every URL in the existing crawl is mapped to its closest-intent equivalent on the new site. The output is a versioned spreadsheet with the old URL, the new URL, the redirect type (301 or canonical-only), and a confidence score. Edge cases (paginated category pages, faceted URLs, parameterised search results, paginated blog archives) get explicit treatment rather than catch-all rules. No URL is mapped via a redirect chain — every redirect is a single 301 hop.
Phase 3 — Schema and hreflang preservation (weeks 3–4)
Every JSON-LD block from the existing site is catalogued and a port specification written for the new build. Where the new templates use a different framework (Next.js, Astro, etc.), the schema-emission helpers are reviewed against current Google rules to ensure the migration is also a chance to clean up any deprecated or invalid blocks rather than carry them forward. Hreflang reciprocity is mapped URL-by-URL across language and region variants.
Phase 4 — Staging audit (week 5)
A full crawl of the staging environment, treated as a production audit. Render-vs-source HTML check on every template (a JavaScript-rendered template that does not return content in the initial HTML response is a ranking risk regardless of how good the visible site looks). Robots and canonical correctness, internal link graph comparison against the existing site, schema validation, hreflang reciprocity check. Any issues here are cheap to fix; the same issues post-launch are not.
Phase 5 — Launch-day execution (week 6)
A scripted launch checklist runs alongside the development team. Robots.txt updated to allow indexing on the new domain, sitemap submitted to Search Console, GSC change-of-address filed (if domain change), redirect smoke tests run on a sample of 200 URLs spanning every URL pattern, render-vs-source check on the live origin, first-batch internal-link crawl. Any regression at this stage is rolled back and fixed before the team breaks for the day.
Phase 6 — Post-launch monitoring (weeks 7–14)
Eight weeks of daily monitoring. Search Console index coverage report, crawl errors, Core Web Vitals field data, rank tracking on every priority query, log file sampling once a week. The first three weeks typically show 10 to 20% short-term volatility as Google processes the redirect map. Any regression that does not resolve inside seventy-two hours is investigated and fixed the same day. By week eight, the new site is performing at or above the pre-migration baseline on all priority queries, or we have a written explanation for why not.
Migration pricing,
scoped to complexity.
Migration SEO is project-priced rather than retainer-priced because the scope and duration are bounded by the engineering work. The price floor is £3,500 for a single-language, single-region migration of up to ~5,000 indexable URLs with no domain change. From there, complexity additions price as follows:
- Domain change: +£1,000 (GSC change-of-address, dual-domain monitoring, redirect maintenance plan).
- Multi-language hreflang: +£1,500 per additional language pair beyond the first.
- Catalogue restructure: +£2,000–£5,000 depending on URL count over 10,000.
- Multi-property consolidation: +£3,000–£6,000 for merging two or more sites into one.
- Recovery engagement (post-failure): bespoke quote, typically £4,500–£12,000 depending on traffic loss and time elapsed.
Most clients also retain us for an eight-week post-launch monitoring window at £950–£1,800/month. That is where 90% of preventable ranking issues surface and get fixed inside hours rather than weeks.
What is migration SEO and when do I need it?
Migration SEO is the discipline of preserving organic search rankings, traffic and pipeline value through any change that alters the structure of your site. Four scenarios qualify: replatforming (e.g. Shopify to Saleor, WordPress to Next.js), full redesigns where URL structure or templating changes, rebrands that change the domain or naming, and consolidations where you merge two or more properties into one. Any of those without a migration SEO plan typically loses 20 to 70% of organic traffic in the first eight weeks. With a plan, the loss is usually under 5% and recovers inside three months.
How long before launch should we engage you?
Six weeks before the planned go-live is the sweet spot. That gives time for a baseline crawl, full URL inventory, redirect mapping, schema preservation review, hreflang audit if you run multi-region, a staging-environment audit, and a pre-launch dry run. We can compress to three weeks if necessary, but at that pace we cut competitive deconstruction and post-launch playbook prep, both of which are valuable. Engage us after launch and we are doing recovery, which is more expensive and slower than prevention.
How do you actually preserve rankings during a migration?
Five disciplines. One: a pre-launch crawl baseline so we know exactly what currently ranks and where. Two: a 1-to-1 URL mapping spreadsheet, every old URL pointed to the closest-intent new URL, with no chain redirects. Three: schema preservation, every JSON-LD block re-emitted on the new URL because lost schema is a slow ranking decay rather than a fast one. Four: a launch-day checklist (robots.txt, canonical correctness, sitemap submission, GSC property changes, hreflang refresh). Five: an eight-week post-launch monitoring window where we watch crawl logs, Search Console data, and rank tracking daily, fixing regressions inside the same day.
What kinds of migrations have you handled?
WordPress to Next.js (most common, usually for performance and developer experience), Magento to Shopify, Shopify to Saleor, Wix to WordPress, custom legacy CMS to headless. Domain changes from .co.uk to .com, brand consolidations where two acquired companies become one site, regional split-outs where one .com becomes a .com plus .co.uk plus .ca. Catalogue restructures where a 40,000-product e-commerce site changes its category architecture. Multi-language hreflang corrections where the previous setup leaked rankings between en-GB and en-US.
Do you only do the SEO part or can you ship the migration?
Both. Most engagements are SEO-only — your developers ship the migration, we own the rankings discipline. We write the URL mapping spec, the redirect rules, the schema migration plan, the GSC playbook, and we monitor post-launch. For Next.js and WordPress builds we can also ship the migration end-to-end through our web design service, which is simpler when SEO and engineering need to be in lock-step. The split is roughly 70% SEO-only, 30% full-build.
What does a migration SEO engagement cost?
Project pricing based on scope. A standard mid-size migration (up to ~5,000 indexable URLs, single language, no domain change) is £3,500. Add complexity — multi-language hreflang, domain change, catalogue restructure, multi-property consolidation — and pricing rises to £6,000–£15,000 depending on URL count and architectural depth. Most clients also retain us for an eight-week post-launch monitoring window at £950–£1,800/month, which is the period when 90% of preventable issues surface.
What happens if rankings drop after launch despite the plan?
Some volatility is normal. Google takes four to eight weeks to re-crawl, re-index, and re-evaluate the new URL set. A 10 to 20% short-term dip in the first three weeks is common, and recovers as Google processes the redirect map. What is not normal is sustained loss after week eight. If we hit that, we pull the post-launch monitoring data, identify the specific failure (commonly: redirect chains we missed, a forgotten canonical, a robots directive change), fix it, and re-submit. Recovery is fastest when the diagnostic happens fast, which is why post-launch monitoring is the highest-leverage line item in the budget.
Can you recover a migration that already went badly?
Yes, and it is more common than the prevention work. The pattern: a developer pushed a redesign live without a redirect map, or with a flat catch-all redirect to the new home page. Rankings collapse 40 to 80% inside two weeks. We do an emergency baseline-versus-current crawl, build the missing redirect map retrospectively, fix the schema, hreflang and canonical mistakes, and submit the corrected sitemap. Recovery is typically 80 to 95% of pre-migration position inside ten to fourteen weeks, depending on how much link equity was burned in the gap.
Verified clients, real migrations, measurable outcomes.
WordPress to Next.js migration on a £4M e-commerce site. Six-week prep, daily monitoring for eight weeks post-launch. Total organic loss across the migration window: 3.1%. Fully recovered by week ten. The redirect map alone was worth the fee.
Shopify to Saleor migration with a catalogue of 12,000 SKUs and a category restructure. Syed flagged six things our developers had not accounted for, including a faceted-URL trap that would have generated 200,000 indexable URLs of duplicate content. Avoided.
Rebrand and domain change from co.uk to .com. Pre-launch baseline, 4,800-line URL mapping spreadsheet, hreflang refresh, and the GSC change-of-address playbook. Less than 5% traffic loss across the change, recovered in seven weeks.
Recovery engagement, not prevention. Previous developer launched a redesign with a flat redirect-to-home rule. We had lost 64% of organic. Hired Syed for emergency recovery. Restored 91% of pre-migration position inside twelve weeks.
Migration coming up? Tell us when.
Send the launch date, the platform you are moving from and to, and the rough URL count. One-day reply, written by Syed.
Ready to protect rankings through your migration?
A short introduction, your site URL, and what you’re trying to achieve. If it’s a fit, we’ll book a 30-minute call.
Free £500 SEO audit included with any web dev or SEO package · no card required