How Long SEO Takes to Recover After a Migration

The question behind this one is usually not academic. Traffic has moved after a migration, and you want to know whether to wait or to panic.

Here is the honest answer: a temporary dip is normal, the shape of the curve tells you more than the depth, and there is a specific report that separates settling from broken. It is not the traffic graph.

The shape of a normal recovery#

Search engines do not re-evaluate your site the moment you launch. They re-crawl it URL by URL, on their own schedule, and each redirected address has to be fetched, followed, and reassessed before it settles at its new home. That process produces a recognisable curve.

Week one and two: movement, and it looks alarming. Google is re-crawling. Some pages have already been reassessed at their new address, some have not, and rankings wobble while the two states coexist. A dip of 10 to 25 percent here is ordinary. Resist the urge to change anything, because you will not be able to tell later which change did what.

Weeks three and four: the turn. Most of your significant URLs have been re-crawled and the redirected addresses start consolidating at their new homes. This is when the curve should begin climbing. If it is still falling at the end of week four, stop waiting and start diagnosing.

Weeks six to eight: settled. For a small or mid-sized site, this is where you should be back at your previous level, sometimes above it if the rebuild also fixed speed and structure. Larger sites take longer for one unglamorous reason: Google has more URLs to work through and only allocates so much crawl to any one site.

The report that actually answers the question#

The traffic graph cannot tell you whether your migration worked. It shows a number going down in both the normal case and the broken case, and by the time the difference is obvious in traffic you have lost weeks.

Search Console's Page Indexing report can tell you, and immediately.

Open it and look at the reasons rather than the totals:

• "Page with redirect" climbing, then falling as those pages move to indexed, is exactly right. That is the migration working through.
• "Not found (404)" climbing is the failure mode. Every URL in that list is an address Google knew, that you moved, and that now leads nowhere. These are not recovering on their own.
• "Excluded by noindex" on pages that should be public means a staging setting shipped to production. This one is urgent and takes minutes to fix.
• "Crawled, currently not indexed" on newly rebuilt pages is usually patience rather than a fault, especially in the first month.

Cross-reference the 404 list against your redirect map. Every entry in it is a row you missed, and each one is fixable in the time it takes to add a redirect.

What separates a dip from a disaster#

Three signals, in the order they are worth checking.

Does the loss have a shape, or is it everywhere? Damage from a bad redirect map is usually concentrated. One section, one URL pattern, one content type. A site-wide, even loss is more often a re-crawl in progress or something unrelated to the migration entirely, like an algorithm update in the same week.

Did the drop start at launch or a few days later? An immediate 50 percent-plus fall on day one is mechanical: a blocked robots.txt, a noindex, a DNS or hosting misconfiguration. Those are the good kind of problem, because they are found and fixed in an afternoon.

Are the pages losing traffic the ones whose URLs changed? If your top-performing page kept its address and still lost rankings, the redirect map is not your problem and you should look at what happened to the page itself during the rebuild. Content that was rewritten during a migration is a common and often unnoticed cause: the page no longer says what it used to rank for saying.

What you can and cannot speed up#

You can help Google find the changes: resubmit your sitemap, use URL Inspection to request indexing on your most valuable pages, and make sure your internal links point at the new addresses rather than relying on redirects to do the work. Internal links still pointing at old URLs is a surprisingly common leftover, and it wastes crawl on hops that did not need to exist.

You cannot buy crawl speed. Recovery pace is largely a function of how often Google already crawls your site, which follows from your size and authority. Anyone offering to accelerate the recovery itself is selling you the wait.

The version nobody wants to hear#

If it has been three months and traffic is flat at a lower level, that is not slow recovery. Recovery finished, and it finished lower.

The causes are almost always one of three, and all three are diagnosable:

1. A section of the site was never redirected. Blog archives and paginated URLs are the usual casualties, because nobody thinks of them as pages.
2. Something is still blocking crawling or indexing. A robots.txt rule or a leftover noindex, quietly costing you everything.
3. The content changed. Pages were tightened or reworded during the rebuild and no longer match the queries they used to win.

None of those improve by waiting, and all of them are visible in Search Console if you look at the indexing report rather than the traffic chart.

The short version#

A dip is normal. A dip that turns upward by week four is working. A dip that keeps deepening is a fault, and the Page Indexing report will name it while the traffic graph is still ambiguous. Plan a migration for four to eight weeks of settling, and never land one three weeks before your peak season.

Worth knowing before you start: this whole recovery period is the reason the migration timeline is longer than the build. If you want to know whether your own site is ready to move without losing this ground, the migration risk score covers the eight gaps that cause it, in about two minutes.