Article

Website Migration SEO Checklist 2026: Protect Rankings Before, During & After Launch

Planning a website redesign, CMS change, domain move or URL restructure? This practical SEO migration checklist explains what to protect before launch, what to verify during deployment and what to monitor afterward so rankings, backlinks, indexation and qualified organic traffic are not lost unnecessarily.

Mahyar Molavi
Written by Mahyar Molavi Manager
Website Migration SEO Checklist 2026
IN THIS ARTICLE Table of contents 21 sections

A new website can look significantly better than the one it replaces and still become an SEO problem.

That is one of the uncomfortable realities of website migration.

The new design may be faster. The CMS may be easier to manage. Navigation may feel cleaner. Content may be rewritten. Development may be technically successful.

Then organic traffic drops.

Pages that previously ranked disappear from Search. Valuable backlinks point through broken URLs. Google starts indexing unexpected versions of pages. Important service pages lose internal links. Old URLs return 404 errors, while the new website looks perfectly normal to everyone reviewing it manually.

The problem is rarely “Google does not like the new website.”

More often, the signals Google understood before the migration were changed faster than they could be transferred or rebuilt.

Google’s own site-move documentation makes this clear: when URLs change, the recommended process includes preparing the new site, mapping old URLs to their new locations, using permanent redirects, updating canonical references and internal links, and submitting the new URLs through an updated sitemap. Temporary Search fluctuations can still happen while Google processes the move.

A safe website migration therefore is not simply a development deployment.

It is a controlled transfer of search signals.

First, understand what kind of migration you are actually doing

“Website migration” can describe several very different projects.

You might move from one domain to another.

You might rebuild a WordPress site using a different theme or development stack.

You might move from WordPress to another CMS.

You may redesign the entire website while keeping the same domain.

Your URL structure may change from something like:

/services/seo/

to:

/seo-consulting/

Or the visible URLs may stay exactly the same while the site moves to a new hosting provider or CDN.

These projects do not carry identical SEO risk.

Google separates migrations with visible URL changes from infrastructure changes where the URLs stay the same. A hosting or CDN migration, for example, follows a different process because search engines do not need to relearn a completely new set of public addresses.

The first migration decision should therefore be:

What is actually changing?

That question sounds simple, but it prevents a common problem: changing far more than the project requires.

Use a “change budget” during migration

One of the most useful principles in migration work is to limit unnecessary simultaneous changes.

Suppose a business moves to a new domain.

At the same time, it changes CMS, rewrites every service page, removes half the blog, changes every URL, redesigns navigation and launches a completely different internal-linking structure.

If traffic falls afterward, what caused it?

The domain move?

Missing redirects?

Content changes?

Navigation?

Canonical errors?

Removed pages?

Changed Search Intent?

You have created too many variables.

Google's Change of Address guidance specifically notes that keeping similar site architecture during a domain migration helps signals transfer more directly. Combining a move with substantial redesign, content and URL changes can require Google to reassess individual pages and may contribute to traffic loss.

This does not mean a migration can never include a redesign.

Often the redesign is the reason for the migration.

It means every additional change should have a reason.

If an existing URL is already descriptive, receives traffic, has backlinks and fits the new structure, there may be no SEO benefit in changing it simply because the new development team prefers a different naming convention.

Preserving what already works can be one of the most valuable migration decisions you make.

Before migration: build a baseline you can compare against

You cannot confidently diagnose a traffic drop after migration if you do not know what was working before it.

Before development reaches launch day, create a clear baseline.

This should include more than a screenshot of total organic traffic.

You need to understand which URLs matter.

A page receiving 20,000 visits per month clearly deserves attention, but traffic is not the only measure of migration value.

A page may receive fewer visits while ranking for a commercially important query.

Another page may have excellent backlinks.

A low-traffic legal or support page may be important for internal navigation.

An older article may still provide internal links to several valuable landing pages.

A good URL inventory combines information from multiple sources rather than relying on the XML sitemap alone. Current 2026 migration guidance from Semrush similarly recommends combining crawl data, Search Console, analytics, backlink data, XML sitemaps and CMS records when creating the pre-migration URL inventory.

This is important because an XML sitemap is not necessarily a complete record of every URL that currently has SEO value.

Some important URLs may no longer be in the sitemap.

Backlinked URLs may still exist.

Legacy pages may continue receiving visits.

Search Console may expose pages that the migration team forgot were indexable.

The old site needs to be understood before it is replaced.

Crawl the old website and save the evidence

Before launch, crawl the existing website.

Do not assume you can reconstruct this information later.

Once the new site replaces it, the original internal-linking structure, status codes, metadata and canonical relationships may no longer be available.

Save the crawl.

At minimum, you want a record of URLs, HTTP status codes, titles, meta descriptions, canonical URLs, indexability, headings, internal links and relevant structured data.

For larger migrations, I would also segment pages by type.

Service pages.

Blog articles.

Products.

Categories.

Portfolio pages.

Location pages.

Language versions.

This makes post-launch comparison significantly easier.

Instead of asking:

“Did traffic fall?”

you can ask:

“Did the service directory lose visibility while informational content remained stable?”

That is a much more useful question.

Decide which URLs stay, move, merge or disappear

This is where migration planning becomes real SEO work.

Each important old URL should have a deliberate outcome.

It stays at the same URL.

It moves to a new URL.

It is consolidated into another relevant page.

Or it is intentionally removed.

A migration should not create hundreds of accidental outcomes.

One of the most damaging patterns is allowing the development team to remove old URLs simply because equivalent templates no longer exist in the new design.

Another is redirecting almost every deleted URL to the homepage.

Google explicitly warns against redirecting many old URLs to one irrelevant destination such as the homepage, because this can confuse users and may be treated as a soft 404.

The redirect destination should normally represent the closest useful replacement.

If:

/services/technical-seo/

becomes:

/seo/technical-seo-consulting/

the mapping is straightforward.

If three thin pages have been genuinely consolidated into one stronger resource, all three may reasonably redirect to that new consolidated page.

But if an old article has no meaningful replacement, sending users to an unrelated service page purely to preserve “SEO value” is usually the wrong approach.

Relevance matters.

Redirect mapping is one of the highest-value parts of the migration

A redirect map is not something to generate five minutes before deployment.

It should be reviewed before launch.

For permanent URL changes, Google recommends server-side permanent redirects where technically possible, including HTTP 301 or 308 redirects.

The ideal mapping is direct:

Old URL → Final new URL

Not:

Old URL → Previous redesign URL → Temporary URL → Final URL

Google can follow redirect chains, but its guidance is to redirect directly to the final destination whenever possible and keep chains short. Chains also add latency for users.

This becomes especially important on websites that have already been redesigned several times.

An article may have changed URL in 2021, again in 2024 and now again in 2026.

If the redirect rules simply keep stacking, the migration inherits years of technical debt.

Before launch, flatten those chains where possible.

The original valuable URL should point to the current final destination.

Do not forget the pages that have backlinks

Migration discussions often focus on current rankings.

Backlinks deserve equal attention.

Imagine an industry publication linked to one of your guides three years ago.

That link may still send authority and referral traffic.

If the migration removes the destination without a relevant redirect, you have broken an asset that another website gave you.

Export important linked pages before migration and make sure their destinations are represented in the redirect map.

After launch, it can also be worthwhile updating especially valuable external links when you have a realistic relationship with the referring website.

Google itself recommends trying to update important external links after a site move, alongside profile links and advertising destinations.

You cannot update every backlink on the internet.

That is why the redirect still matters.

Staging should protect the migration, not create a new SEO problem

A staging environment is essential for serious redesign work.

It is also a frequent source of migration mistakes.

The staging site should not become an accidentally indexable copy of production.

But the opposite mistake can be even more damaging:

noindex protection is added during development and accidentally remains active when the website goes live.

Google lists leftover noindex rules and robots.txt blocks among common site-migration problems that can prevent the new website from being properly indexed.

This is why migration QA needs two separate states.

Before launch: prevent the staging environment from appearing in Search.

After launch: verify that the production pages intended for Search are crawlable and indexable.

Never assume this changed correctly because the deployment succeeded.

Check it.

Test the new site as if Google had never seen it before

A visual QA is not an SEO QA.

Developers may confirm the navigation works.

Designers may confirm every breakpoint.

Content editors may confirm the copy is correct.

SEO still needs to ask different questions.

Can crawlers reach the important pages?

Do canonical tags reference the final production URLs?

Are any canonicals still pointing to staging?

Are multilingual hreflang references updated?

Do internal links point directly to new URLs rather than old URLs that redirect?

Are important pages missing from navigation?

Have previously indexable pages accidentally become blocked?

Does the XML sitemap contain only the URLs that should now be indexed?

Google specifically recommends updating self-referencing canonicals, hreflang annotations where applicable, and internal links to the new URLs before or during a move.

This is not cosmetic cleanup.

These signals help Google understand the new structure.

Update internal links instead of relying on redirects forever

A 301 redirect is valuable.

It does not mean internal links should continue pointing to the old URL forever.

Consider:

Article → Old service URL → 301 → New service URL

That works.

But:

Article → New service URL

is cleaner.

After migration, your website should increasingly behave as if the old URLs no longer exist.

Update navigation, contextual links, breadcrumbs, canonical references, image references and other internal paths where necessary.

Redirects are the safety net for old links and external references.

They should not become the permanent architecture of the new website.

This is also the right moment to review Internal Linking quality.

A redesign often changes more than URLs.

Cards disappear.

Footer links change.

Blog templates change.

Navigation becomes smaller.

Those design decisions can alter how much internal authority reaches important pages.

A migration QA should therefore compare relationships between pages, not only URL status codes.

Prepare the new XML sitemap before launch

The new sitemap should represent the new site.

Not the history of the old one.

URLs that now redirect should not remain as the primary URLs in the current sitemap.

The sitemap should contain the canonical, indexable destination URLs you actually want search engines to discover.

Recent 2026 technical guidance also flags redirecting URLs inside XML sitemaps as a problem because the sitemap should point search engines toward the live destination rather than a URL that immediately redirects.

After migration, submit the updated sitemap in Google Search Console.

Google includes this as part of its recommended site-move process.

The sitemap will not repair a broken migration.

It helps Search discover the new structure more efficiently.

Launch day: treat deployment as a controlled handover

A migration should have an SEO launch checklist, not only a development deployment checklist.

At launch, several things need to become true at almost the same time.

The production site becomes available.

Permanent redirects activate.

Canonical tags reference production.

Temporary indexing protections disappear.

The final robots.txt rules become active.

The new sitemap is available.

Internal links resolve correctly.

Tracking continues working.

Search Console can inspect important production URLs.

A useful launch structure looks like this:

Migration stagePrimary SEO goal
Before launchPreserve evidence, map URLs and test the future site
LaunchMake the new site crawlable and transfer old URLs correctly
First 72 hoursFind critical failures before they spread
First 30 daysMonitor indexing, traffic and ranking movement
30–90 daysResolve persistent issues and compare performance by segment

The first crawl after deployment is especially valuable.

It can reveal errors that did not exist on staging.

Broken production routes.

Incorrect environment variables.

Canonical domains left unchanged.

Redirect rules behaving differently on the production server.

Missing assets.

Unexpected 404s.

You want to discover those problems before Google repeatedly discovers them.

If the domain changes, use Search Console correctly

A domain migration requires an additional signal.

After the new domain is live and the redirects are working, Google's Change of Address tool can be used to tell Google about a move from one domain or subdomain to another.

But this tool is not required for every migration.

Google says not to use Change of Address for:

HTTP to HTTPS moves.

www to non-www changes on the same domain.

Moving URLs between paths on the same domain.

Hosting changes where the public URLs remain unchanged.

That distinction matters.

Search Console tools should reflect what actually changed.

The Change of Address tool does not replace redirects either.

The redirects must already be in place.

Keep redirects much longer than launch week

Removing migration redirects after one or two months is an unnecessary risk.

Google currently recommends keeping redirects in place for as long as possible and generally at least one year so it has enough time to transfer signals, recrawl URLs and reassign links pointing to the old addresses.

For users, keeping important redirects even longer may make sense.

Old bookmarks exist.

Backlinks remain.

Documents may contain old links.

Emails may still be opened years later.

The migration does not end because your development team has closed the project ticket.

The old URL history remains part of the web.

Expect some movement after migration

One mistake is treating every ranking fluctuation during the first few days as evidence the migration failed.

Another mistake is assuming a large decline is normal and doing nothing.

Both reactions are dangerous.

Google says Search visibility may temporarily fluctuate while a site move is processed, and that small to medium sites may take several weeks for most pages to move while larger websites can take longer.

That gives us useful context.

Temporary volatility is possible.

But volatility still needs investigation.

If an important Page 1 URL disappears and its replacement returns noindex, that is not something to “wait out.”

If hundreds of old URLs return 404 because a redirect configuration failed, waiting for Google to settle is not a strategy.

The difference between normal processing and a technical failure comes from monitoring the actual signals.

Monitor by page group, not just total traffic

Sitewide organic traffic can hide migration problems.

Suppose traffic is down 8%.

That number alone tells you very little.

Maybe Blog traffic is stable while high-value service pages lost 40%.

Maybe the English section recovered immediately but the German URLs have incorrect hreflang.

Maybe desktop performance is normal while mobile conversions fell after the redesign.

Segment the data.

At minimum, compare:

commercial landing pages,

informational content,

directories or categories,

countries,

devices,

languages,

brand versus non-brand queries,

and strategically important URLs.

Search Console should be checked more frequently after launch than during normal operation.

Google also warns that its crawlers may hit the new site more heavily during migration because requests to old URLs can redirect into the new infrastructure in addition to normal crawling. Server capacity therefore matters too.

A migration can be technically correct and still struggle if the new infrastructure cannot reliably serve increased crawling.

Watch for the migration problems that repeatedly cause damage

Most migration disasters are not mysterious algorithm events.

They are predictable implementation failures.

A recent 2026 technical SEO case highlighted issues after a migration including orphaned priority URLs, incorrect menu mappings, thousands of redirect chains, thousands of 404 errors, parameter URLs being indexed, canonical problems and missing hreflang annotations.

That list is useful because it reflects a larger lesson.

Migration SEO is mostly about preserving consistency.

Old URL → correct destination.

Internal link → final URL.

Canonical → current indexable URL.

Sitemap → current 200-status URLs.

Language annotation → correct equivalent.

Navigation → important page.

Tracking → same business event.

When these relationships break, traffic can break with them.

Do not judge the migration only by traffic recovery

Preserving SEO does not mean the goal is to recreate the old site forever.

A migration may also be an opportunity to improve.

Perhaps the old architecture had several overlapping pages.

Maybe important services were buried.

The old WordPress implementation was painfully slow.

Content templates were weak.

Mobile UX hurt conversion.

Analytics did not measure qualified enquiries.

Those are valid reasons to migrate.

The objective is not simply:

“Keep every old signal exactly the same.”

It is:

Preserve what has value, deliberately replace what needs improvement, and make every change traceable.

This is the difference between an SEO-led migration and a redesign that hopes Google figures everything out afterward.

The migration is not complete when traffic returns

The technical move may finish on launch day.

SEO migration work continues afterward.

Once indexing stabilises, compare the new site with the baseline.

Did important pages recover?

Did some pages improve?

Are previously weak pages now performing better?

Did crawl waste increase?

Did conversion rates improve?

Did the new information architecture make commercial pages easier to discover?

Are new URLs earning backlinks rather than relying permanently on old redirects?

A good migration should eventually leave the website in a stronger state than before.

Not simply alive.

A practical migration decision framework

Before approving a migration, I would ask four questions.

What business problem requires this change?

If there is no strong reason to migrate, unnecessary technical risk is difficult to justify.

Which existing SEO signals already have value?

Identify the URLs, rankings, links, content and internal relationships worth protecting.

Which changes must happen together?

Separate necessary migration changes from improvements that could safely happen later.

How will we know if something went wrong?

Define the baseline, monitoring and ownership before launch.

If nobody knows who checks redirects, Search Console, analytics and crawl errors after deployment, the migration plan is incomplete.

Final Website Migration SEO Checklist

Before launch, you should be able to answer yes to the important questions.

The old website has been crawled and its important URLs are documented.

Organic landing pages, rankings and backlinks have a baseline.

Every important URL has a planned outcome.

Changed URLs have relevant permanent redirects.

Redirect chains have been minimised.

Canonical references use the final URLs.

Internal links point directly to the new destinations.

Staging protections will not reach production.

robots.txt does not accidentally block important sections.

Multilingual annotations have been updated where necessary.

The production XML sitemap contains the correct canonical URLs.

Analytics and conversion tracking have been tested.

Search Console access is ready.

The team knows what will be checked immediately after launch.

And most importantly, SEO is part of the migration plan before the old website disappears.

A website migration does not need to preserve every ranking perfectly from the first hour.

Some fluctuation is normal.

But preventable technical mistakes should not be confused with normal fluctuation.

The safest migrations are usually not the ones with the most sophisticated SEO tools.

They are the ones where developers, designers, content teams and SEO agree on exactly what is changing, what must be protected and how the new site will replace the old one.

If you are planning a redesign, WordPress rebuild, CMS migration, domain change or major URL restructure, the SEO work should begin before development reaches launch day.

A migration review should identify valuable URLs, redirect requirements, indexation risks, internal-linking changes and measurement requirements while those decisions are still inexpensive to change.

Once the old site has been replaced and traffic is already falling, the same work becomes recovery rather than prevention.

Mahyar Molavi
About the author

Mahyar Molavi

Manager

I have been developing websites and creating content for over ten years, and this is just the beginning...

EMAIL[email protected]
WORK TYPERemote / Project Based
RESPONSE TIMEUsually within 24–48 hours

Start a collaboration

Collaboration type (optional)

I’ll respond as soon as possible.

Start a collaboration

Have a project in mind? Let’s talk about it

If you need help designing, developing, optimizing or growing a digital product, share the first details and we can discuss the most practical path.

Available for new projects