SEO Migration Checklist: How to Switch Stacks Without Losing Rankings
A practical SEO migration checklist for switching CMS platforms without losing rankings: from pre-migration benchmarks to post-launch website monitoring.

TL;DR
Whether you're migrating a client's website or overseeing an enterprise-level replatform, a structured SEO migration checklist ensures you protect organic rankings, preserve link equity, and keep every critical element optimized, from redirects to metadata. In a competitive search landscape, a site migration doesn't have to mean a traffic dip. Done right, it's a strategic opportunity to strengthen your site's performance from the ground up.
Why You Need an SEO Migration Checklist
SEO migration checklist is must have for any type of replatforming, or website migration. It's the difference between rankings that hold and traffic that quietly disappears. You can have great content and strong backlinks and still lose rankings if the technical layer breaks. A CMS migration is where that layer is most at risk.
Switching CMS platforms is one of the highest-risk things you can do to a website's search performance. If you migrate from a legacy monolithic CMS to a modern headless setup, or from an expensive SaaS platform to a self-hosted alternative: the SEO risk is the same. Not because the migration process is technically hard, but because the damage is invisible until it isn't.
Website rankings don't drop in a moment. Your positions drop three-four weeks later, when Google finishes reprocessing what has changed.
We'll show three main phases: what to measure before you touch anything, what breaks SEO silently during the migration, and what to watch for in the 28 days after you ship.
![]()
Search engines spend years building a picture of your website: which pages exist, what they're about, how they relate to each other, how fast they load. That accumulated understanding drives your rankings.
When you [migrate a site from one CMS to another], you're asking search engines to rebuild that picture from scratch. If the signals they relied on (your URLs, your page structure, your metadata) have changed or disappeared, rankings drop while they recalibrate. For many sites, that recalibration takes weeks. For some, months.
The part that's easy to get wrong: the damage often isn't visible right away. A page that silently lost its metadata, a redirect that wasn't quite mapped correctly, a configuration file carried over from the staging environment, none of these look like errors. The site works. Users can navigate. But search engines are receiving different signals than they were before, and rankings respond to that eventually.
A study of 892 website migrations found average recovery time, when SEO wasn't treated as a technical priority, was 523 days. Nearly 17% of sites never recovered to pre-migration levels.
Most of this is preventable. It requires treating SEO as part of the migration plan from day one, not a QA step at the end.
If you're planning a CMS migration and want to understand what a well-run project looks like end to end.
Phase 1: Before you migrate, establish your baseline
The most important thing you can do before touching a single file is document where you stand today. Without SEO migration checklist, you have no way of knowing whether the website migration helped, hurt, or left things unchanged.
Capture your current rankings and traffic
Export everything carefully, at least 12 months of data from Google Search Console and Bing Webmaster Tools: which pages receive clicks and impressions, which queries drive traffic, where your pages rank on average. Do the same from your analytics platform for organic sessions and conversions.
These numbers are your benchmark. After you ship, any significant deviation is worth investigating. A 10-15% temporary dip is normal. A sustained 30% drop means something went wrong.
Save every URL on your current site
Before anything changes, get a complete list of every URL currently indexed by Google. Blog posts, product pages, category pages, landing pages, everything.
This list becomes the foundation for your redirect map: the document that ensures every old URL points to the right new location after migration. There's no way to redirect what you haven't catalogued.
In our experience, skipping this step is one of the most consistent patterns in migrations that struggle to recover.
Pay particular attention to URL parameters. Legacy systems often use query parameters, things like ?category=shoes or ?page=2. If the new platform names those parameters differently, every old indexed URL becomes a dead end. It's quiet, easy to miss in testing, and genuinely destructive to rankings.
Document your current SEO setup
Take stock of everything currently configured for SEO: page titles and descriptions, any structured data markup (the code that enables rich results like star ratings or FAQ dropdowns in search), and how your pages link to each other. If your site targets multiple languages or regions, document how that's handled.
None of this transfers automatically to a new platform. It needs to be deliberately rebuilt (that proves the necessity of seo migration checklist again, and again), and you can only do that if you know what was there in the first place.
Phase 2: During migration, what breaks silently
This is where the real risk lives. The issues below don't produce error messages. The site keeps working. But they change how search engines understand and rank your content, and the effects compound.
| Element | What breaks | Why it's invisible | Fix |
|---|---|---|---|
| Redirects | Page authority disappears | Site navigates normally, no errors shown | Map every old URL before migration; verify 301s post-launch |
| Metadata | Titles and descriptions missing or empty | Pages render correctly for users | Verify both CMS fields and frontend rendering layer independently |
| Canonical tags | Authority split across duplicate URLs, or full deindexation | No visible symptom until rankings drop | Ensure canonicals point to live domain, not staging, before ship |
| Structured data | Rich results (stars, FAQs, cards) stop appearing | Page looks identical before and after | Rebuild markup in new templates; validate with Google's Rich Results Test |
| robots.txt | Entire site removed from search index | Site fully accessible to users | Check file immediately post-deploy; name a responsible person |
| HTML structure | Search engines misread page topic and hierarchy | Page appears visually unchanged | Audit H1–H3 order in new templates against pre-migration baseline |
| URL parameters | Old indexed URLs become dead ends | New platform silently renames parameters | Document all legacy parameters; include in redirect map |
The last two rows, HTML structure and URL parameters, are worth adding because both are called out specifically in the article and are genuinely underappreciated failure points. The URL parameters one in particular is flagged as "quiet, easy to miss in testing, and genuinely destructive" — it earns a row.
Redirects: the highest-stakes item
When URLs change during a migration, and they almost always do, redirects are how you tell search engines: this page moved, here's where it lives now. A permanent redirect (301) transfers the page's accumulated authority to the new URL.
Missing or incorrect redirects are the single most common cause of post-migration ranking loss. If an old URL returns an error instead of redirecting, the authority it built up over years effectively disappears.
Every URL on your old site needs either a redirect to its new location, or a deliberate decision to retire it. There's no safe middle ground.
Metadata: titles, descriptions, and more
Page titles and meta descriptions don't live in your content, they live in the code that wraps it. When you move to a new platform, that code gets rebuilt, and metadata doesn't carry over automatically.
Different CMS platforms handle metadata differently. In WordPress, it's managed through a plugin. In a headless setup, SEO fields need to exist in the content model inside the CMS, and the frontend needs to read and render them server-side. When you switch platforms, these two layers don't migrate together automatically, and there's no guarantee the new CMS even has the same field structure as the old one.
This is a double point of failure that generic migration guides don't talk about. You can have perfect SEO fields in your new CMS and still ship pages with empty titles if the rendering layer isn't configured to use them. And you can have perfect rendering logic that silently falls back to empty strings if the content model is missing fields. Both need to be verified independently.
Pages can end up with empty titles, identical titles across the whole site, or descriptions that existed in the old setup, but don't exist in the new one. Search engines use this information to understand and rank pages. Losing it, even temporarily, affects visibility.
Canonical tags: the duplicate content trap
A canonical tag tells search engines which version of a page is the official one. It prevents duplicate content issues when the same content is accessible at multiple URLs.
New platforms often generate pages at multiple URLs by default: with and without trailing slashes, with different parameter combinations. Without correct canonical tags, search engines split a page's authority across several versions instead of concentrating it on one.
There's a specific migration trap worth flagging: during development, canonical tags often point to the staging server rather than the live domain. If this isn't corrected before you ship, you're telling Google that every page on your live site is a duplicate of a page on a server users can't access. The result is deindexation, over weeks.
Structured data: the invisible layer
Structured data is the code behind rich search results: FAQ dropdowns, review stars, recipe cards, event listings. If your site currently appears with any of these in search results, that's powered by structured data markup.
This markup lives in page templates, not in content. When templates are rebuilt on a new platform, structured data often doesn't come across — and there's no visible sign that anything is wrong. The page looks identical. The information search engines use to generate rich results is gone.
This matters more now that AI-powered search is growing. More on that below.
HTML structure: what Google actually reads
The heading hierarchy of a page (H1, H2, H3), are signals to search engines what the page is about and how the content is organized. When templates are rebuilt from scratch, this structure can change in ways that aren't obvious visually but matter to crawlers.
A page can look identical before and after migration while its heading structure has been flattened, duplicated, or reordered. That decision compounds over time as rankings shift to reflect how Google now understands the page.
robots.txt: the configuration that can erase your site
robots.txt tells search engine crawlers which pages they're allowed to visit. During development, it's standard to block all crawlers so the staging environment doesn't surface in search results.
The problem: that configuration sometimes ships to production. A robots.txt that blocks everything on your live site will, over time, remove your entire site from search results. No errors. No warnings. No symptoms, until your traffic drops.
Verifying this file is correct after you ship takes thirty seconds. It should be a named step on every deployment checklist, with a named person responsible.
Phase 3: After you ship, monitoring and recovery
The work doesn't end when you ship. This is where the most important work begins.
The first week
Run a full crawl of the live site immediately after shipping and compare it against your pre-migration baseline. Look for missing redirects, pages with empty or incorrect titles, canonical tags pointing to staging, heading structure changes.
Check Google Search Console's Coverage report for 404 errors. A spike means redirects are missing. Fix these fast, every day they stay broken is a day authority is leaking away from pages that earned it.
Verify that robots.txt is correct. Do this first.
The 28-day window
Some SEO signals are updated slowly. Core Web Vitals, Google's metrics for page loading speed and responsiveness, are measured using real user data collected over a rolling 28-day window. A rough ship can lock in poor performance scores for up to a month, even if the underlying issues are fixed the next day.
This is the argument for measuring performance on staging before you ship. Problems caught there don't leave a trace in Google's data. Problems caught after do, for 28 days.
During this window, monitor weekly:
- Organic traffic compared to your pre-migration baseline, broken down by site section
- Keyword rankings for your most important terms
- The Coverage and Core Web Vitals reports in Google Search Console
A 10-20% temporary dip is normal as search engines process the changes. A sustained decline beyond that, or a decline concentrated in a specific section of the site, points to something specific worth investigating.
The second crawl
At or around the 28-day mark, run another full crawl and compare it against both your pre-migration baseline and your day-one post-launch crawl. Through this data you will understand, how and where things have stabilized, and what, if anything, still needs attention.
A note on AI search
Traditional SEO remains the primary driver of organic discovery, and the biggest area of risk during a migration. But the landscape is changing.
AI-powered search, e.g. Google's AI Overviews, Perplexity, ChatGPT, now drives a meaningful and growing share of traffic across most industries. In some verticals, AI-generated answers already account for more initial discovery than traditional results.
One risk that's specific to this migration path: most AI crawlers don't render JavaScript. Testing of major AI systems, ChatGPT, Perplexity, Claude, and others, has consistently shown that none of them execute client-side rendering.
If you're migrating from a traditional server-rendered CMS to a headless setup and any of your key pages end up as JavaScript-dependent SPAs, those pages are effectively invisible to AI discovery systems, even if Google indexes them without issue.
This is a new failure mode specific to the headless transition. Googlebot has gotten much better at rendering JavaScript over the years. AI crawlers haven't. The fix is the same as for traditional SEO: use SSR or SSG for content that needs to be discoverable, but the reason to care is different, and most migration checklists don't mention it.
Structured data now carries more weight than it used to. In traditional Google search, structured data was the path to rich results, useful, but optional for most sites.
In AI-powered search, it's one of the primary ways AI systems understand and classify what your content is about. A website migration that loses structured data loses visibility in both systems simultaneously.
Everything else in this SEO migration checklist: redirects, metadata, URL structure, monitoring, applies equally to both traditional and AI search. Get the fundamentals right, and you're covered on both fronts.
SEO Migration Checklist: Final Steps
If CMS migration done right, it wiill be invisible to your content team. A migration done wrong shows up in your analytics, GSC, or SEMrush, four weeks later. The gap between the two is almost always in the preparation.
If you need a full technical audit, or SEO migration services, share your case and let's get in touch with our headless CMS agency. We'll review your setup carefulle, and help you migrate without losing the rankings you've already earned.






