SearchAuditPro
LoginStart Free Trial

Site Migration SEO: How to Build a Redirect Map and Verify It After Launch

Site Migration SEO: How to Build a Redirect Map and Verify It After Launch

By Joe Whyte · 11/22/2026

Most migration problems are not caused by the move itself. They come from a redirect map that was built in a hurry, never tested, and never checked after launch. A redesign, a platform switch, a domain change or even a URL structure cleanup can all undo years of accumulated search equity if the old URLs do not point cleanly to the right new ones.

This is the process we use and recommend: snapshot the old site, build the map deliberately, test before launch, and verify in the weeks after. None of it is exotic, but skipping any step tends to be expensive.

What counts as a migration

People often think of migrations as only domain changes. In practice, any of these changes URLs at scale and deserves the same care:

  • Moving to a new domain or subdomain
  • Switching from HTTP to HTTPS
  • Changing CMS or ecommerce platform
  • Restructuring categories or folders
  • Removing trailing slashes, changing casing or altering parameters across the site
  • Merging two sites into one

If URLs change, redirects are needed. If content also changes substantially, expect some ranking movement even when everything is done correctly. The goal is to keep that movement small and temporary.

Step 1: Snapshot the old site before anything changes

You cannot map what you have not listed. Before the build starts, capture a complete record of the current site. This is your baseline for every comparison later.

Collect, for every indexable URL:

  • The URL, status code and canonical target
  • Title tag, H1 and meta description
  • Internal links pointing to it
  • Organic landing page data from your analytics and Search Console
  • External backlinks pointing to it

A full crawl gives you the first four. For sites up to 1,000 pages, a page-by-page audit in a tool like SearchAuditPro can produce that inventory in one pass, and the backlink data can be matched against it so you know which URLs have external links worth protecting. Export it and store it somewhere safe. You will want it again after launch.

Also record your current rankings for your most important keywords. Without a pre-migration baseline, you cannot tell a normal fluctuation from a real loss.

Step 2: Decide which URLs need a redirect

Not every old URL deserves a destination. Sort the inventory into three groups:

  1. Pages that continue to exist under a new URL. These get a one-to-one redirect.
  2. Pages that are being merged or replaced by a closely related page. These redirect to the most relevant replacement.
  3. Pages with no equivalent. If they have no traffic and no links, let them return a 404 or 410. If they have value, find the closest relevant page, or reconsider removing them.

The temptation is to send everything without a clear match to the homepage. Avoid this. Google may treat mass redirects to an irrelevant page as soft 404s, and users who land on a homepage when they expected a specific product or article tend to leave.

Step 3: Build the redirect map

A redirect map is a simple table: old URL in one column, new URL in the next. The discipline is in how you fill it.

Use rules where patterns exist, manual entries where they do not

If /blog/2023/post-name becomes /articles/post-name across hundreds of pages, one pattern rule is cleaner than hundreds of lines. For pages whose slugs or structure changed individually, write manual entries. Most real migrations need both.

Redirect to the final destination, not the next hop

Redirect chains happen when A goes to B and B goes to C. Each hop adds latency and risk. When you build the map, point every old URL straight to its final live URL, and update older redirects from previous migrations at the same time. Sites that have been redesigned more than once often carry chains nobody remembers creating.

Use permanent redirects

Use 301 or 308 for permanent moves. A 302 or 307 signals a temporary change, and while Google can eventually treat long-lived temporary redirects as permanent, there is no reason to rely on that.

Keep the same content intent

A redirect works best when the destination answers the same question as the source. If an old guide is being replaced by a shorter, different page, expect the ranking to be reassessed. That is not necessarily a problem, but it is not the same as a clean move.

Do not forget non-HTML assets and parameters

PDFs, images that earn links, and URLs with tracking or filter parameters that have been indexed can all matter. Check Search Console and your backlink data for assets that receive visits or links.

Step 4: Prioritize by value

With thousands of URLs, you need to know where an error would hurt most. Rank the old URLs by organic entrances, referring domains and conversions. The top slice of that list should be checked by hand, one by one, before launch. The long tail can be validated in bulk.

Pages with strong backlinks deserve special attention. If an external site links to an old URL and the redirect fails or lands somewhere irrelevant, that link stops helping you. Matching your backlink data to your map is one of the most useful checks in the whole project.

Step 5: Test on staging, then test again at launch

Before launch, run the old URL list against the staging environment, or against the redirect rules if staging is not possible. Check for:

  • Every old URL returning a 301 or 308 to the intended target
  • Final targets returning 200, not another redirect or an error
  • No chains and no loops
  • Redirects preserving the right path, not stripping everything to the root
  • Staging not being indexable by accident, and, just as important, the live site not inheriting a noindex or a robots.txt block from staging

That last point has caused more painful launches than almost anything else. Make it an explicit item on the launch checklist: confirm the robots.txt file and meta robots tags on the live site right after go-live.

On launch day, crawl the new site and test the old URL list again. Confirm that:

  1. Canonical tags point to the new URLs, not the old ones
  2. The XML sitemap lists only new, indexable, 200 status URLs
  3. Internal links point directly to new URLs instead of relying on redirects
  4. Hreflang, structured data and Open Graph tags carry the new URLs
  5. Analytics and Search Console are installed and working on the new setup

If the domain changed, add and verify the new property in Search Console and use the change of address tool where it applies.

Step 6: Monitor for weeks, not days

After launch, the work shifts to verification. Expect some volatility. What you are looking for is whether it settles and whether any pattern points to a specific failure.

In the first week

  • Re-crawl the new site and compare it with your pre-launch baseline. Look for new 404s, broken internal links, missing titles and pages that lost their canonical or became noindex.
  • Check Search Console for crawl errors, a rise in excluded pages, and the status of the submitted sitemap.
  • Review server logs or crawl stats if you have access, to see whether Googlebot is requesting the old URLs and getting the right responses.

In the following weeks

  • Track your priority keywords against the baseline you saved. A modest dip that recovers is normal. A sustained drop on a specific section suggests a mapping or content problem in that area.
  • Compare organic landing page traffic, section by section, with the old site.
  • Watch for old URLs that still appear in search results long after the move, which can point to redirects that are not being followed or crawled.
  • Check that valuable backlinks now resolve to live pages. Where you can, ask the most important linking sites to update their links to the new URLs.

A repeat audit every couple of weeks during this period is cheap compared with the cost of discovering a problem two months late. Scheduling a recurring crawl and rank tracking on the same keyword set makes the comparison much easier.

Common mistakes to avoid

  • Redirecting everything to the homepage. It looks tidy and loses relevance.
  • Leaving internal links pointing at old URLs. Redirects should catch outside traffic, not serve as your own site navigation.
  • Changing too much at once. If you can avoid redesigning content, changing URLs and switching domains in a single release, do so. Fewer variables make problems easier to diagnose.
  • Removing redirects too soon. Keep them in place for as long as old URLs have links or traffic. In practice, that often means indefinitely.
  • Launching without a baseline. If you did not record the before state, every later question becomes guesswork.

A short pre-launch checklist

  1. Full crawl and ranking baseline saved
  2. Old URLs sorted into keep, merge and remove
  3. Redirect map built with direct, one-hop targets
  4. Top pages by traffic and links checked manually
  5. Staging tested, indexing directives confirmed on live site
  6. Sitemap, canonicals and internal links updated to new URLs
  7. Search Console and analytics ready
  8. Post-launch crawls and rank checks scheduled

A migration will rarely be perfectly neutral, but a careful map and a disciplined verification routine keep the cost low and the cause of any problem easy to find. The work is mostly in the preparation, and the preparation is mostly lists, comparisons and patience.