Reading Search Console after a relaunch: which dips are normal

The fortnight after a website switchover is when owners lose their nerve. Search Console shows impressions dipping, a few pages slip down the results, and someone suggests reverting. Often nothing is wrong: the movement is Google working through the move. Some of the time something is broken, and the job of the first thirty days is to tell the two apart quickly enough to fix the broken part while the old URLs still carry weight.

This is the relaunch counterpart to the post-launch 30 day measurement plan. The principle carries over: fix the instruments first, read a small number of things properly, and spend the month reaching one decision. The daily and weekly schedule is in the website relaunch runbook; this guide covers how to read what Search Console shows.

What does Google say is normal after a site move?

Google's own guidance sets the expectation: temporary fluctuation in ranking, a few weeks for most pages of a small to medium-sized site to move, longer for larger ones. Plan the first month around that timing rather than around the hope of a flat line.

The Google Search Central guide to site moves with URL changes says to expect temporary fluctuation in site ranking during the move, that a small to medium-sized website can take a few weeks for most pages to move, and that Google will temporarily crawl the new site more heavily than usual after a migration. It also says to keep redirects for as long as possible, generally at least one year. Google does not define the bands, but an industrial site with a few hundred product and application pages is likely to count as small to medium, so a dip that lasts two or three weeks and then recovers is inside the expected range.

The same guide tells you where the move will show: the indexing graphs will reflect it, and the performance reports will start showing new URLs gaining impressions and clicks. Those are the two places to watch, in that order.

Which movements are expected, and which signal a broken move?

The test is whether traffic is moving or disappearing. Traffic that leaves an old URL and arrives on its new equivalent is a healthy move; traffic that leaves and arrives nowhere is a fault with a cause you can find.

The last row is the one that does the most damage and takes the least time to fix. It is on the first-hour list in the relaunch day checklist, and it is worth checking again on day two because deployments sometimes overwrite it.

How should you read the Page indexing report in week one?

Read it for errors on specific URLs, not for the total. In week one the indexed count barely moves, and Google's help notes it can take a week or so to start crawling and indexing a new page or site.

The Page indexing report help page defines each status. The ones that matter after a relaunch:

• Page with redirect. An old URL that redirects. This is what you want old URLs to show, and Google notes such a URL will not be indexed.
• Redirect error. Chains that are too long, loops, overly long redirect URLs or bad URLs in the chain. Each one is a page whose ranking is not being passed on.
• Not found (404) and Soft 404. Fine for pages you removed on purpose with no replacement. A fault for any product or category page the map should have covered.
• Discovered and Crawled, currently not indexed. Common and not always a problem, but on a relaunched site it often points at product pages whose specifications moved into a download rather than the page.
• Blocked by robots.txt, and pages excluded by a noindex tag. On the new site, these are almost always deployment mistakes.

Use the URL Inspection tool on your twenty most valuable old URLs and their new equivalents. The report itself cannot be filtered by a single URL, and inspection confirms each redirect target page by page. When you fix a class of errors and start validation, allow for Google's note that validation typically takes up to about two weeks.

Why should you keep both sitemaps submitted?

Because the old sitemap prompts Google to recrawl old URLs and find the redirects, which is how the move gets processed. Google's site move guide says to submit both, and that the redirect warnings on the old one can be ignored.

In the Sitemaps report, the new sitemap should show its URLs being read. Compare the count of URLs submitted against the count of new pages you expect to exist: a large difference usually means the sitemap generator skipped a section or still includes staging addresses. If the domain itself changed, submit a Change of Address for the old property, which the same guide recommends for domain and subdomain changes.

How do you compare old and new URLs in the Performance report?

Take your top fifty old URLs by clicks from the three months before launch, pair each with its new equivalent from the redirect map, and check week by week that clicks lost on one appear on the other. This comparison shows directly whether each page's search traffic is being passed on.

Export the old pages before launch day. In the Performance report, group by Pages and use a page filter for each pair, or export weekly and match the pairs in a spreadsheet. Read clicks, impressions and average position together. A healthy pair shows the old URL falling and the new one rising, with the combined total roughly holding. A broken pair shows the old URL falling and the new one flat. That pattern usually traces to a redirect that sends the visitor to a category or the home page instead of the matching product, a new page that lost the specification text that ranked, or a new URL that is not indexed at all.

Ignore the last two or three days of data when you compare. The Performance report help page notes that the newest data can be preliminary and might still change, shown as a dotted line on the chart.

The redirect type matters here too. Google's guidance on redirects explains that permanent redirects show the new target in results while temporary ones keep the source page showing. A relaunch built with temporary redirects by default will show old URLs holding on longer than they should.

Why does Core Web Vitals wait until week four?

Because the report runs on a rolling 28 day window of real visits from the Chrome UX Report, and only indexed URLs appear. Read it in week four, when its window covers the new site rather than a blend of old and new.

The Core Web Vitals report help page explains that URLs are grouped by similar experience, that a group needs a minimum amount of data to appear, and that validation after a fix runs a four week monitoring period. For an industrial site with modest traffic, many groups will fall back to the origin level or not show at all. Treat an empty report as a lack of data, not as a pass, and check key templates with lab tools in the meantime.

What decision should you make at day thirty?

One of three: the move is on track and you keep monitoring monthly, specific pairs need repair, or the redirect map needs a full audit. Write the choice down with the evidence, as in the post-launch plan.

1. On track. Most top pairs show traffic transferring, the Page indexing errors are explained, and impressions are back near or above the pre-launch baseline. Keep the redirects, keep the old sitemap until old URLs drop out, and move to monthly checks.
2. Repair specific pairs. A handful of pairs are broken, usually for the same reason. Fix the redirect targets or restore the missing page content, request indexing for the fixed pages, and recheck in two weeks.
3. Full redirect audit. Many pairs are broken, or errors cluster across a whole section. Crawl every old URL from the pre-launch export, record the response and final destination, and fix the map before anything else changes on the site.

Reverting to the old site is rarely the right call at day thirty. It restarts the move from the beginning and leaves Google with two sets of signals to reconcile.

For a structured outside view of how the new site performs across indexing, redirects and page quality, the EbizIndia SEO audit report covers the same ground in more depth. If the decision behind the relaunch is still being questioned internally, what a rebuild can cost you on SEOtop10 sets out the risks an owner signs up to.