Old Domain Still Showing in Google? How to Fix It After a Migration
If your old domain still shows in Google, it is either a stale result that clears as Google recrawls your redirects, or Google choosing the old host because signals conflict. How to tell which, and fix it.
· 15 min read
If your old domain is still showing in Google after a move, one of two things is happening: Google has not yet recrawled and processed all your redirects, which is normal for weeks after a migration, or Google is still choosing the old host as canonical because your signals conflict. Inspect one old URL in Search Console. If it 301-redirects in a single hop to the exact new URL and Google's selected canonical is the new URL, wait: Google says a small to medium-sized site can take a few weeks for most pages to move, and larger sites take longer. If the old URL returns 200, chains through several redirects, or Google still selects it as canonical, fix the conflicting signal (redirects, canonical tags, sitemaps or internal links) before requesting a recrawl. (developers.google.com)
A site migration can appear healthy for months and then reverse: in one reported case, more than 5,000 new-domain URLs became “Crawled – currently not indexed” while impressions shifted back to the old host. When an old domain still showing in Google causes that pattern, the payoff is not simply getting a new crawl—it is identifying the conflicting sitewide signal that is making Google select the wrong host.
The Reddit case behind this guide involved a domain rename completed in early 2026, a recovery to roughly 5,000 daily impressions on the new domain, and an apparent reversal four months later. Old-domain search results redirected visitors correctly to matching new URLs, but Google Search Console attributed visibility to the old property. That is a domain-canonicalization problem until proven otherwise, not a reason to immediately noindex the new site.
Why Google can show an old domain after a migration
A 301 redirect is a strong canonical signal, but it is one signal among several. Google’s migration guidance recommends mapping old URLs to their corresponding new URLs, redirecting the old URLs, updating URL references, and monitoring both hosts. Its canonicalization documentation also identifies redirects and rel="canonical" as strong signals, while sitemap inclusion is a weaker one.
That matters because a migration can send mixed instructions. For example, oldsite.com/guides/widget may 301 redirect to newsite.com/guides/widget, while the new page’s canonical tag still names oldsite.com/guides/widget. Google then has two contradictory signals for the same document.
An old host can also remain discoverable through:
- XML sitemaps containing old-domain URLs.
- Internal links, navigation, breadcrumbs, image links, or hreflang annotations pointing to the old host.
- A CMS, CDN, plugin, or server rule generating canonicals with the old domain.
- Redirect chains, intermittent 302 responses, or redirects that all send pages to the new homepage.
- External links that Google continues to recrawl, particularly where the old host returns a live 200 response for some URLs.
The original report is notable because old URLs redirected correctly when clicked. That reduces the likelihood of a total redirect failure, but it does not confirm that every one of the 5,000-plus mappings, canonical tags, sitemaps, and internal references agrees. A representative manual check can miss a template-specific error affecting thousands of paginated, category, tag, or post URLs.
Separate stale old-domain results from wrong-host indexing
Not every old-domain result is an emergency. A search result can temporarily display an old URL while Google recrawls and consolidates signals, especially after a large move. If the old result redirects in one hop to its exact new equivalent and the new URL is the Google-selected canonical, the result may be a stale migration remnant.
A genuinely wrong-host pattern has stronger evidence. Search Console may show the old domain receiving the majority of impressions, while equivalent new URLs are excluded as “Crawled – currently not indexed” or reported as duplicates. In URL Inspection, Google may select the old URL as canonical despite the new page declaring itself canonical.
Use this simple classification before changing directives:
- Stale but correct: Old result appears, old URL 301 redirects once to its exact new counterpart, and URL Inspection identifies the new URL as Google-selected canonical.
- Incorrect canonical selection: Old URL is indexed or selected as canonical while the new equivalent exists. Look for conflicting canonical tags, indexable old pages, old sitemap entries, or internal linking.
- Redirect implementation issue: Some old URLs return 200, 302, 404, a chain, a loop, or a blanket redirect to the homepage instead of the equivalent new page.
- New-site indexability issue: The new page returns 200 but has
noindex, a crawl block, thin or changed content, a soft-404-like template, or another reason Google does not retain it in the index.
This distinction prevents a damaging reaction: blocking or noindexing the new domain because the old domain is visible. That action removes the destination Google should be choosing.
Inspect both domains in Google Search Console
Google Search Console should contain verified properties for both the old and new domains. For a domain migration, the Change of Address tool is used after the new site is live and old URLs are redirected. Google describes the tool as a way to emphasize crawling and indexing of the new site and to prefer it when determining canonical pages; it supports, but does not replace, a technically sound migration.
Start with the two properties side by side. Compare the same date range and isolate a representative set of pages: homepage, high-traffic articles, category pages, products, pagination, and URLs reported as excluded. A 20-page sample spread across templates is more useful than inspecting 20 blog posts generated from the same template.
What to record from URL Inspection
For each old/new URL pair, record the following fields from the indexed inspection result:
- Indexing status and the stated exclusion reason.
- User-declared canonical on the new URL.
- Google-selected canonical on the new URL.
- Last crawl date and crawl-allowed status.
- Whether the old URL is redirected and the final destination reached by Google.
The URL Inspection tool can show Google’s indexed view, test the live URL, and accept a request to crawl an individual URL. Crucially, the live test cannot predict Google’s eventual canonical selection; the indexed result is where the Google-selected canonical appears.
For a deeper explanation of interpreting exclusions such as a sudden concentration of “Crawled – currently not indexed,” see this Google Search Console indexing guide. In a migration, that status should prompt a host-signal audit before an agency treats it as a generic content-quality issue.
Test representative old and new URL pairs
A domain migration should have a one-to-one map wherever equivalent content exists. Test the full HTTP journey for at least 20 representative URL pairs and then crawl the complete old-domain URL inventory. Browser testing alone is insufficient because browser cache, extensions, and client-side routing can hide server behavior.
For each old URL, the expected route is:
https://old-domain.example/path/ → 301 → https://new-domain.example/path/ → 200
The final new URL should return a self-referencing canonical such as:
<link rel="canonical" href="https://new-domain.example/path/">
Flag any route that contains more than one hop, changes protocol or host unnecessarily, drops query parameters that matter, or lands on an irrelevant destination. Google treats permanent server-side redirects as a canonical signal, but a broad rule sending 5,000 old articles to a new homepage is not a meaningful page-to-page migration.
Also test URL variants. Check http and https, www and non-www, trailing-slash variations, uppercase legacy URLs if applicable, and parameters that once generated crawlable pages. A hidden alternate host can become the source of a surprising old-domain resurgence.
Crawl for migration-wide host mismatches
A migration reversal is usually found faster in a crawl than in individual URL Inspection tests. Crawl the new domain and, where technically possible, the old-domain URL list. Configure the crawl to collect status codes, redirect destinations, canonical tags, meta robots, robots directives, internal outlinks, sitemap URLs, and hreflang values.
Audra can be used as a local-first crawl layer for part of this work: it follows every internal link, flags broken links and redirect chains longer than one hop, and checks each page's canonical tag and robots meta, without moving a client’s crawl data to a subscription platform. It audits up to 1,000 pages per report, so for a full old-to-new mapping of thousands of URLs, a crawler with a URL-list mode is the better tool. The useful output is a mismatch list, not merely a total count of indexable pages. If budget is the constraint, the best free SEO audit tools include free crawlers and Google's own inspection tools that cover most of these checks.
Prioritize these crawl filters
A migration audit should isolate at least six filters:
- New-domain pages with canonicals pointing to the old domain.
- New-domain internal links pointing to the old domain.
- Old-domain URLs returning 200 instead of 301.
- Old URLs with redirect chains, loops, 302s, or irrelevant destinations.
- XML sitemap URLs on the old host or sitemap URLs that redirect.
- Indexable new URLs carrying
noindex, blocked resources, or inconsistent hreflang targets.
For large sites, review template clusters rather than fixing a random list of URLs. If every /blog/ page has an old canonical but product pages do not, the source is likely a content template, plugin setting, cached deployment artifact, or reverse-proxy rule—not an individual editor mistake.
The Audra vs Screaming Frog XML sitemap audit guide is also useful when the sitemap itself needs to be reconciled with canonical, indexable destination URLs. During a migration, a sitemap is a consistency check: it should advertise only the preferred new-host URLs that return 200.
Align redirects, canonicals, and indexability signals
The repair is usually straightforward in principle: every important signal should point toward the new domain. The work becomes difficult when different systems own different signals—web server, CMS, SEO plugin, CDN, translation platform, and sitemap generator.
Use 301 redirects for old URLs that have moved permanently. Keep the old domain under control and preserve those redirects rather than allowing it to expire or serve replacement content. Google’s own site-move documentation specifically treats incorrect redirects, noindex or robots blocks, and outdated sitemaps as migration troubleshooting areas.
Use self-referencing canonicals on preferred new pages. A new article at new-domain.example/post-a/ should ordinarily canonicalize to itself, not to the historical old URL. Canonicals are not a substitute for redirects: a canonical is a preference signal for duplicate or very similar pages, whereas the redirect moves users and crawlers to the replacement URL.
Do not place a blanket noindex directive on the new domain. Inspect templates for accidental noindex caused by staging settings, tag archives, page builders, CMS privacy controls, or deployment environment variables. Google must be able to crawl a URL to see its noindex; blocking it in robots.txt can prevent Google from discovering the directive.
The old-domain migration target generally should not be noindexed if it is supposed to redirect. A redirecting URL is not a page that needs an HTML canonical and noindex stack. First ensure its server response is the correct permanent, one-hop redirect.
Remove old-host discovery paths
Even flawless 301 redirects compete against an old host that the new site keeps advertising. Remove old-domain URLs from every sitemap index and XML sitemap, then submit the new-domain sitemap in the new Search Console property. A sitemap is a discovery hint, not a command, but submitting old URLs tells Google to keep requesting the obsolete addresses.
Audit internal links in HTML, JavaScript-rendered navigation, structured-data URLs, image URLs, RSS feeds, email archives where practical, and CMS-generated breadcrumbs. A single old-host link in a footer can create links across every page; a crawl will show its scale immediately.
Multilingual sites require an extra pass. hreflang annotations must use final, canonical new-domain URLs and reciprocal language/region relationships. Google’s internationalization guidance supports hreflang through page annotations or sitemaps; a mix of old and new hosts in a language cluster can create unclear targeting signals.
This same consistency principle matters beyond classic Google results. A complete brand discovery audit for search, AI, and social can reveal where historical domains still appear in linked profiles, citations, feeds, or brand references. Those sources may not determine Google’s canonical choice directly, but they can sustain user confusion and continued old-host discovery.
Use re-indexing and removal tools for the right job
A URL Inspection request is appropriate after correcting a significant individual page or after validating representative repaired templates. Google says recrawling can take from several days to several weeks, and a request does not guarantee immediate inclusion or any particular ranking outcome. It should not be treated as a sitewide repair for 5,000 inconsistent URLs.
Request recrawls for a small, strategic sample: the homepage, a high-value page from each fixed template, and a few URLs that previously showed the wrong Google-selected canonical. Monitor whether the indexed data changes after Google processes the repair.
The Google Search removal tool is for urgent visibility control, not migration canonicalization. It can temporarily hide a verified page, directory, or site from results, but the underlying technical signals still need correction or the result can return. It is reasonable when legal, privacy, reputational, or brand-confusion risk requires a rapid temporary hide of the old domain.
Avoid using temporary removal simply because an old-domain result redirects correctly. Hiding the old result may reduce immediate confusion, but it does not solve the reason Google selected the old host. First repair redirects, canonicals, internal references, and sitemaps; use removal only when the business need is genuinely urgent.
Expected timing and what not to do
There is no fixed number of days for Google to replace every old URL with its new-domain counterpart. Google's own guidance is that a small to medium-sized website can take a few weeks for most pages to move, and larger sites take longer, and that redirects should be kept for as long as possible, generally at least 1 year. The timeline varies by site size, crawl frequency, number of changed URLs, server reliability, redirect quality, and whether signals agree. Google advises monitoring both old and new URLs during a move, while its recrawl guidance gives a broad range of days to weeks for a requested crawl—not a promise of instant reindexing.
A four-month delay before symptoms appear is possible when Google revisits a deeper set of legacy URLs or detects a signal that was intermittently wrong. Examples include an expired CDN rule, an old sitemap regenerated by a scheduled job, a plugin update restoring the former site URL, or a redirect rule that fails on older paths.
Do not take these shortcuts:
- Do not redirect every old URL to the homepage. Map equivalent content where it exists; return an appropriate 404 or 410 only where no replacement exists.
- Do not block the old domain in
robots.txtwhile relying on Google to process redirects. Google needs to crawl old URLs to see their redirect targets. - Do not add canonicals from old pages to new pages while old pages return 200. Make the move unambiguous with permanent redirects.
- Do not repeatedly request indexing for thousands of URLs. Correct the shared cause and use requests to validate a sample.
- Do not judge success by one
site:query. Compare Search Console impressions, clicks, indexed canonical selections, sitemap coverage, and crawl results over time.
Agency validation checklist for the client report
A client-ready report should show the before-and-after evidence rather than simply state that redirects were “checked.” For a reported 5,000-URL issue, include counts by template and a sample of exact URL pairs.
Use this validation checklist:
- Both old and new domains are verified in Google Search Console.
- The Change of Address request was submitted for an actual domain or subdomain move, after redirects went live.
- At least 20 representative old URLs return one-hop 301 redirects to exact new equivalents.
- The final new URLs return 200 and declare self-referencing new-domain canonicals.
- URL Inspection shows the new URL as the Google-selected canonical for repaired samples.
- New-domain pages have no accidental
noindex, password wall, robots block, or staging directive. - Old-domain URLs are absent from new XML sitemaps, and sitemap entries resolve to canonical 200 URLs.
- Internal links, hreflang, structured data, feeds, and navigation use the new domain.
- Redirect chains, 302s, loops, 200-status old pages, and homepage catch-all redirects are resolved.
- Search Console performance is compared by host weekly until the new domain becomes the consistent visibility owner.
For agencies managing both traditional rankings and answer-engine discovery, attach a separate baseline for branded mentions and citations. The AI search visibility measurement guide explains why that evidence should be tracked separately from Google Search Console data. A successful domain migration should preserve a coherent destination across both technical crawl signals and the places where a brand is surfaced.
FAQ
Why is Google still indexing posts under my old domain?
Google may still be processing a valid migration, or it may be receiving stronger or more consistent old-host signals. Check whether old URLs return one-hop 301 redirects, whether new pages canonicalize to themselves, and whether Google-selected canonicals in URL Inspection name the new host. Old sitemaps, internal links, 200-status old pages, and broken template rules are common causes.
How do I tell Google that my site moved to a new domain?
First implement page-to-page 301 redirects from the old domain to the new domain, update canonicals, internal links, and XML sitemaps, then use Search Console’s Change of Address tool for a domain or subdomain move. Verify both properties and monitor each host. The tool supports the move, but it cannot override conflicting redirects, canonicals, or indexability problems.
Should old-domain URLs use 301 redirects, canonicals, or noindex tags?
For a permanent domain migration, old URLs should normally use server-side 301 redirects to their closest new equivalents. New pages should use self-referencing canonicals on the new host. Do not rely on canonical tags instead of redirects, and do not blanket-noindex redirecting old URLs. Use noindex for pages intentionally excluded from search, not as the default domain-migration mechanism.
How long does it take Google to replace old-domain pages with new-domain URLs?
There is no guaranteed timeline. Google says a small to medium-sized site can take a few weeks for most pages to move, larger sites take longer, and an individual requested URL may be recrawled in days to weeks, but a complete migration depends on crawl frequency, site size, server responses, URL mapping, and consistent signals. Track the trend in Search Console and validate canonical selections on representative pages. A sudden reversal after several months warrants a technical audit rather than passive waiting.
Can I remove old-domain URLs from Google Search immediately?
Search Console’s Removals tool can temporarily hide verified URLs, directories, or a site from Google Search when there is an urgent need; Google says removal requests last for about 6 months. It is not a permanent migration fix: results can return if the underlying pages and signals remain unchanged. For normal domain moves, correct the redirects, canonicals, internal links, and sitemaps so Google has durable reasons to select the new domain.
Why does my old website still show up on Google after I moved?
Usually because Google has not finished recrawling the old URLs and processing their redirects. Old results can keep appearing for a few weeks on a small to medium-sized site, and longer on large ones. If an old result redirects in one hop to its exact new page, it will normally be replaced. If old URLs return 200, redirect to the homepage, or Google still selects them as canonical in URL Inspection, something on the site is telling Google the old host is still the right one.
How long should I keep redirects from the old domain?
Google's site-move guidance says to keep redirects for as long as possible, generally at least 1 year, so it can transfer signals and reassign links that other sites still point at your old URLs. Removing redirects early is a common reason old-domain URLs reappear or links stop counting.
Sources
- https://www.reddit.com/r/SEO/comments/1x2wm8u/google_search_randomly_started_indexing_posts/
- https://developers.google.com/search/docs/crawling-indexing/site-move-with-url-changes
- https://support.google.com/webmasters/answer/9370220?hl=en-GB
- https://support.google.com/webmasters/answer/9012289?hl=en
- https://developers.google.com/search/docs/crawling-indexing/consolidate-duplicate-urls
- https://developers.google.com/search/docs/crawling-indexing/301-redirects
- https://developers.google.com/search/docs/crawling-indexing/ask-google-to-recrawl
- https://support.google.com/webmasters/answer/7479439?hl=en-CA
- https://developers.google.com/search/docs/crawling-indexing/remove-information