Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

Checklists

Website Migration Checklist

A website migration checklist for SEO and tracking: inventory and benchmarks, redirect mapping, staging checks, and launch-day verification.
Checklistintermediate8 min readUpdated

When to run this checklist

Open it at the start of any project that changes URLs, the domain, the protocol, the platform, the templates or the hosting, and keep it open until the monitoring group is complete. The "before" group belongs in the first planning meeting, not the week before launch, because the inventory and the benchmark cannot be taken after the old site is gone and the redirect map takes longer than anyone budgets. Run the staging group each time staging is rebuilt, since a fix in one template can reintroduce a staging directive in another. Run the launch-day group in order, on the day, with the engineers in the room. The after-launch and monitoring groups are then run weekly, and the list is closed only when organic traffic and rankings have returned to the benchmark or the loss has been explained. A redesign that keeps every URL still needs the staging, launch and monitoring groups; it is the "before" group that shrinks. When the move is part of a change of business name, the rebrand rollout tutorial places this checklist inside the wider sequence of ads, profiles, listings and email that change with it. Whether the new templates convert better than the old ones is a separate question from whether the move kept its search visibility, and the guide to evaluating a website redesign sets out the template tests and comparison groups that answer it.

Before you start

  • A full crawl of the current site, exported with status codes, canonicals and titles, plus the URLs that the crawl cannot see: pages found only in server logs, in GSC or in the backlink profile. The crawling and indexing guide explains why those hidden URLs matter.
  • Search Console and analytics access on both the old and the new site, including the property for the new domain if there is one, created before launch so it has history from the first day.
  • A redirect map owner who holds the spreadsheet of old URL to new URL and signs off every change to it; a map edited by several people is a map with gaps.
  • Staging access for the crawler and for the tag container's preview mode, so that both the SEO lines and the tracking lines can be checked before anything is public.
  • A launch window outside peak trading, with engineering available for the rest of the day and a rollback plan written down.
  • The tracking container and the analytics property, so that the GA4 tag, consent and conversion events move with the site rather than being rediscovered a week after launch.

The checklist

Phases are sequential and the groups do not overlap, so tick the "before" group completely before staging work starts and do not open the launch-day group until the staging group is clean. Where a line names a sample, such as testing redirects, the sample should include the pages that earn the most traffic and links, not a random selection. A line that cannot be ticked because the platform does not support it, such as a single-hop redirect on a host that chains them, is a finding for the owner and a risk to record, not a line to skip.

0/31

Before

Staging

Launch day

After launch

Monitoring

If an item fails

Before launch, a failed line is a reason to delay, and the cost of delay is almost always lower than the cost of launching with the gap. A redirect map with unmapped URLs, a staging site that is already indexed, a tag container that does not fire on the new checkout: each is cheaper to fix on staging than to diagnose on a live site that is also losing rankings. On launch day a failed line is a rollback decision, made against the plan written beforehand rather than in the moment. After launch, failed lines are triaged by traffic loss: redirect gaps on pages with links and traffic first, then indexing problems, then tracking, then everything else. A migration also consumes Crawl Budget while the engine reprocesses every URL, so a large site should expect the recovery to take longer than a small one even when every line passes. Tracking failures at any phase go to the tracking verification checklist, and the data quality guide explains how to tell a measurement break from a real loss when the two arrive on the same day.

Group that failedWhere the fix is explained
BeforeThe site migrations guide and site architecture for a map that changes the structure
StagingCrawling and indexing, structured data, the canonical tag entry and hreflang for multilingual sites
Launch dayThe launch-day section of the migrations guide and Google's site move documentation in the sources list
After launchInternal linking and the technical SEO audit checklist for the full post-launch crawl
Tracking at any phaseThe tracking verification checklist and marketing data quality
Illustrative example — not a real result

A software company's domain move

A B2B software company moves from a hyphenated domain to a shorter one and replatforms at the same time, against advice. The "before" group passes except for the freeze: the content team publishes 14 rewritten product pages in the final fortnight, so the content comparison on staging flags 14 changed pages that nobody can attribute to the migration or the rewrite. Staging fails on redirects: a sample of 200 URLs from the map shows 31 chaining through the old domain's trailing-slash rule before reaching the new page. The launch is held for five days while the rule is fixed. On launch day everything passes except the conversion event on the new demo form, which is repaired within two hours from the tag container. Six weeks later organic traffic sits at 94% of the benchmark, with the gap entirely on the 14 rewritten pages; the migration itself held.

To order the checklist's page-level tasks by value, use the migration risk walk-through for the SEO traffic estimator, which ranks pages by conversions at risk.

When the move is complete and organic visits hold but fewer visitors convert, the page on a conversion rate that dropped after a redesign works through the tracking, proof, form, speed and message changes a relaunch can introduce.

Common mistakes

  • Taking the benchmark from memory. Nobody exported rankings or traffic by URL before launch, so the loss is argued about rather than measured.
  • Mapping by pattern and never checking the exceptions. The rule covers most URLs; the handful it misses are the campaign landing pages with the external links.
  • Treating staging checks as a one-off. Staging is rebuilt three times before launch and the indexing block returns on the second rebuild.
  • Testing redirects by opening the homepage. The homepage redirects; the deep pages with the traffic chain, loop or return a not-found page.
  • Moving the site and forgetting the tags. The new templates have no container, or the old container with triggers built for the old page structure, and conversions vanish for a week.
  • Watching the wrong report. Total traffic looks fine because paid and direct are unchanged; the organic loss is visible only when the channel is isolated and compared by landing page.
  • Closing the project at launch. The monitoring group is where migrations are saved or lost, and it needs an owner for months, not days.

The technical SEO audit checklist is the list to run on the new site once the monitoring period ends, and the checklists library holds the rest.

Terms used on this page

Where this page is used

Part of these learning paths

Also related

Sources

  1. Site moves with URL changes — Google Search Central

Related