Domain Migration Indexing Delay: A Four-Month Triage Guide
A practical four-month diagnostic workflow for separating normal domain-move reassessment from redirect, canonical, discovery and indexing failures that need intervention.
· 15 min read
Four months after a domain move, the example behind this guide had only 64 indexed pages while roughly 2,000 remained unindexed: about 1,000 marked “Crawled - currently not indexed” and another 1,000 “Discovered - currently not indexed.” A domain migration indexing delay can be normal in part, but those proportions call for page-level diagnosis—not a vague decision to wait.
Google does not provide a definitive “migration complete” milestone in Search Console. Its Change of Address tool helps Google prioritize the destination and transfer signals, but it does not replace accurate permanent redirects, clean canonical signals, crawlable internal links, or a usable XML sitemap. At four months, the practical goal is to identify which missing URLs are still being reassessed and which are caught in a repeatable technical or content-quality pattern.
Four months is a triage point, not an automatic failure
A domain move can take months, especially where Google must recrawl a large number of old and new URLs and reassess their canonical versions. Google’s Change of Address request is intended to support a domain or subdomain move for a limited period; it is not a dashboard that confirms every URL has transferred successfully. Google’s own site-move documentation also advises keeping redirects in place for as long as possible, generally at least one year.
That means four months is not proof that the migration is permanently broken. It is, however, too late to treat a situation such as 64 indexed URLs out of roughly 2,064 known pages as ordinary turbulence without evidence. The migration described in the original Reddit post had permanent 308 redirects, updated canonicals and internal links, a submitted sitemap, and no stated robots.txt block—yet many previously successful pages had not re-entered the index.
The correct framing is:
- Normal reassessment: the new domain’s indexed-page count rises over time, important URL groups are appearing, Googlebot continues to crawl both domains, and the old-to-new mappings are consistently understood.
- Systemic failure: the same templates, directories, language folders, or page types repeatedly show exclusions; redirects resolve incorrectly; canonicals point elsewhere; or important pages receive little internal discovery.
- Mixed outcome: the highest-value pages migrate but a long tail of thin, duplicate, faceted, paginated, or orphaned pages does not. This may be acceptable only if those pages were never intended as indexable canonical URLs.
The distinction matters because mass-submitting indexing requests or rewriting thousands of pages before understanding the pattern can add work without correcting the real blocker.
Start with a migration evidence set, not a sample of URLs
The first job is to build one dataset that can answer a simple question for every important old URL: where does it lead now, and what does Google think the new URL is? Use the old domain’s historical organic landing pages, XML sitemap URLs, crawl exports, server logs where available, and backlink landing pages as the starting list.
Create a spreadsheet or database with at least these columns:
- Old URL
- Intended new URL
- Final destination after redirects
- HTTP status and redirect type
- Redirect hops
- New-page canonical URL
- New-page robots directive or X-Robots-Tag
- Present in the new XML sitemap
- Number of internal links to the new URL
- Search Console Page indexing status
- URL Inspection result and last crawl date
- Old-domain clicks/impressions before the move
A representative sample is useful for manual debugging, but it cannot establish whether 1,000 excluded URLs share a pattern. Segment the full list by directory, template, content type, and commercial importance. For example, a site with /city/, /service/, and /blog/ URL groups should report indexing rates separately for all three—not as one sitewide percentage.
This approach is also the fastest way to spot a migration map that looks complete in theory but has quiet gaps in practice. A missing rule for trailing slashes, uppercase legacy paths, query-string URLs, a subdomain, or the non-www hostname can strand hundreds of historic URLs.
For a broader mapping process, Audra’s website migration audit guide explains how to compare old and new URL inventories locally rather than relying on a limited sample from a crawler or Search Console.
Segment Search Console statuses before changing pages
Open the Page indexing report in the new-domain Search Console property, export the affected URL examples, and join them to the migration evidence set. Google limits the examples shown for each reason, so the report should guide investigation rather than be mistaken for a complete URL inventory.
Discovered - currently not indexed: solve discovery and crawl priority
“Discovered - currently not indexed” means Google knows about the URL but has not crawled it at the time of reporting. After a domain migration, that can point to a weak discovery path, a very large set of low-priority URLs, server-capacity concerns, or simply a crawl queue that has not reached that group yet.
Run concrete checks:
- Confirm the exact canonical URL is in the new XML sitemap, not an old-domain sitemap or a redirected URL.
- Confirm important hub pages link to it using normal HTML
<a href>links. Google uses links both to discover pages and as a relevance signal. - Check that the page is linked from the migration target’s category, service, location, or editorial architecture—not only from an XML sitemap.
- Review Search Console Crawl stats and server logs for Googlebot requests and recurring 5xx, 429, timeout, or blocked responses.
- Verify all old host variants redirect:
http,https,www, non-www, and any meaningful legacy subdomains.
Do not treat a sitemap submission as a command to index every listed URL. It gives Google a discovery source. Indexability and selection still depend on what Google finds after crawling.
Crawled - currently not indexed: solve selection signals
“Crawled - currently not indexed” is a different problem. Google fetched the page, but it is not currently included in the index. The label does not specify one root cause or prove a sitewide quality judgment. It does mean the next checks must focus on whether the page is a clear, valuable, canonical version of content Google wants to retain.
For each affected template, compare 10 to 20 URLs against indexed equivalents. Look for duplicate or near-duplicate copy, parameter variants, boilerplate-heavy location pages, empty listings, thin category pages, soft-404-like content, conflicting canonicals, and pages whose primary content is absent from rendered HTML. If important pages were copied almost exactly from the old domain, that alone is not a mistake during a site move; the question is whether Google can unambiguously see the new URL as the replacement rather than as an alternate or low-value duplicate.
Test redirects as one-to-one replacements
Permanent server-side redirects are the core migration signal. Google specifically recommends 301 redirects for canonical old pages in its Change of Address guidance, while its broader redirect documentation recognizes server-side redirects as the appropriate mechanism for a moved site. A working 308 is a permanent HTTP redirect, so changing every 308 to a 301 purely because of the code is not a sensible first response; test the actual behavior and mapping instead.
For every priority old URL, the preferred result is one hop to the closest equivalent new URL:
old service page → matching new service page
Not:
old service page → old homepage → new homepage
or:
old city page → generic category page → new city page
A mass redirect to the home page is often a poor substitute for a page-level migration. It weakens relevance, gives users a worse destination, and makes it harder for Google to understand that the old page has a direct successor.
Check the redirect map for these failures:
- Chains: old URL → intermediary URL → final new URL. Remove unnecessary hops.
- Loops: a rule on the new domain sends a URL back to the old host or between hostname variants.
- Soft redirects: JavaScript navigation, meta refreshes, or interstitial pages where a server-side redirect should be used.
- False 200 responses: old URLs return a live page with a canonical tag instead of redirecting.
- Wrong targets: many unrelated URLs converge on a generic destination.
- Lost assets or alternate hosts: migrated subdomains, PDFs, localized directories, and www/non-www versions are omitted.
Keep the old domain online solely to serve redirects. It should not remain a second, crawlable copy of the site with self-referencing canonicals and live duplicate pages. The old host can continue to appear in Google’s results during the handover, but the expected direction is for the new domain to increasingly become the selected canonical destination.
If an old domain remains dominant despite correct redirects, compare the old and new pages in URL Inspection and revisit the exact map. The related guide on old domains still showing in Google covers the practical checks for a migration that has not fully handed over in search results.
Check canonical, robots and rendering signals on both sides
A successful redirect does not override every contradictory signal. For a migrated page, the intended new URL should normally return 200 OK, be indexable, and declare a self-referencing canonical that exactly matches the preferred final URL. Its canonical must not point back to the old domain, to a staging hostname, to an HTTP variant, or to a category page.
Inspect three versions of each priority case:
- The old URL: Does it return one permanent redirect to the intended new URL?
- The live new URL: Does live testing show that Google can fetch it, render it, and access required resources?
- Google’s indexed view: Does URL Inspection show the Google-selected canonical, user-declared canonical, crawl date, and indexing status expected for the page?
A live test only shows whether Google can access the current page. It does not mean Google has already recrawled, selected, or indexed it. That difference is crucial after a migration: a technical fix can be live today while the Page indexing report still reflects an earlier crawl.
Also inspect the HTTP headers as well as the HTML. A noindex directive can be sent through an X-Robots-Tag, and therefore be invisible to someone who checks only the page source. Robots.txt is primarily a crawl-management mechanism; it is not a reliable way to remove a page from Google’s index. More importantly in this case, blocking a page before Google sees a noindex can prevent Google from processing that directive.
For JavaScript sites, compare rendered content with the source and the live test. If the core text, links, canonical tag, or page status only appears after a script fails or a client-side route changes, fix that before expecting broad recovery.
Repair sitemap and internal-link discovery at the same time
Submit a new-domain XML sitemap in Search Console and ensure it contains only final, canonical, indexable URLs that return 200. Exclude redirected old-domain URLs, noindex pages, error URLs, parameter duplicates, and canonicalized alternates. Search Console’s Sitemaps report can confirm whether Google fetched and processed the file, but sitemap inclusion is not a promise of indexing.
Then audit internal links. A migration frequently updates templates but misses editorial links, navigation components, breadcrumbs, hreflang annotations, pagination, HTML sitemaps, or links embedded in JavaScript configuration. Each old-domain internal link sends an unnecessary redirect signal and makes the new site’s architecture less direct.
Prioritize these repairs:
- Update all internal links to final new-domain canonical URLs.
- Link high-value affected pages from relevant category, service, and location hubs.
- Repair orphan pages that are only listed in XML sitemaps.
- Remove internal links to redirecting URLs and redirects within navigation.
- Ensure pagination, hreflang, structured-data URLs, Open Graph URLs, and canonical tags use the new host consistently.
A local crawl makes this substantially easier. Audra can crawl the old and new sites, identify response codes, canonicals, robots directives, inlinks, performance and accessibility issues in one desktop audit, allowing an agency or site owner to keep the evidence locally rather than pay repeatedly to rerun a cloud crawl.
Use Search Console tools for validation, not reassurance
Verify ownership of both the old and new properties. In the old property, use Settings → Change of Address to confirm the request was submitted to the correct destination after the redirects were live. Google says the tool asks it to emphasize crawling and indexing the new site, forwards signals, and prefers the new site when choosing canonical pages. It should be viewed as one supporting signal, not as a pass/fail certificate.
Use URL Inspection on a deliberate sample: the homepage, top 20 historic organic landing pages, 10 affected URLs from each major folder, and several old URLs that should redirect. Record the last crawl date, index status, user-declared canonical, Google-selected canonical, robots accessibility, and any fetch error.
Use the Performance report to compare old- and new-domain clicks, impressions, and top queries before and after the move. A fall in old-domain impressions alongside a rise in equivalent new-domain URLs is evidence of migration progress. A fall in both, particularly for formerly valuable landing pages, signals that mapping, index selection, content rendering, or a broader quality problem deserves priority.
Do not use a site: query as the primary measurement. It can be a quick spot check, but Page indexing, URL Inspection, sitemap status, crawl data, and performance trends provide more actionable evidence.
Prioritize recovery by business value and repeated failure patterns
The fastest recovery plan is not “fix every unindexed URL equally.” Start with pages that had rankings, clicks, conversions, backlinks, or strategic importance before the migration. A 20-page service or product set that previously generated leads deserves deeper inspection before thousands of archive pages.
Week 1: remove migration contradictions
- Export historic high-value old URLs and build the old-to-new map.
- Correct wrong, chained, looping, generic, or missing redirects.
- Fix new-page canonicals,
noindexdirectives, robots blocks, server errors, and staging references. - Submit a clean canonical-only XML sitemap and repair internal links from major hubs.
- Check Manual Actions and Security Issues in Search Console.
Weeks 2 to 4: improve the pages Google has crawled but excluded
For the “Crawled - currently not indexed” cohort, work by template. Make each important page clearly distinct and useful: retain migrated topic coverage, restore any text or structured details lost in the move, provide descriptive titles and headings, and make navigational paths obvious. Avoid mass AI rewrites or superficial word-count expansion; those can alter formerly successful content without solving canonical ambiguity or weak internal linking.
Request indexing sparingly for pages that have materially changed and are commercially important. Google notes that recrawling may take days to weeks and that a request does not guarantee inclusion. The scalable fix is improving the underlying signals, not submitting hundreds of individual requests.
For situations where URLs are indexed but receive no meaningful visibility after the technical work, see page indexed but no impressions for the next diagnostic layer.
Know when to escalate rather than wait
Waiting is reasonable when crawl dates are recent, the count of indexed important pages is steadily improving, and no repeated technical defect appears. Escalate the investigation when the same status persists across valuable pages after corrected signals have been live long enough for recrawling, or when Search Console identifies access, canonical, manual-action, security, or server problems.
The original four-month example is a useful threshold because the pattern is extreme: most known pages are outside the index and the unindexed set is split almost evenly between “discovered” and “crawled.” That suggests two parallel tracks may be needed—improved discovery and crawl paths for one cohort, then canonical/content/index-selection analysis for the other.
A migration should not be judged by whether every legacy URL appears in Google. It should be judged by whether the canonical pages that matter to users and the business have clear replacements, clean signals, stable accessibility, and recovering search performance. When an audit is organized at URL level, the work becomes a prioritised engineering and content problem rather than an indefinite wait for a tool status to change.
FAQ
Why are most pages still not indexed four months after a domain migration?
Four months can still fall within a migration’s reassessment period, but most important pages remaining unindexed is not a reason to simply wait. Separate URLs by Page indexing status, then test redirect destinations, canonicals, robots directives, sitemap inclusion, internal links, server responses and rendered content. A repeated pattern by folder or template usually reveals whether the issue is systemic.
How can a normal domain migration delay be separated from a technical indexing problem?
A normal delay usually shows positive movement: new-domain impressions and indexed important URLs rise while old-domain visibility declines. A technical problem shows stable exclusions, redirect errors, conflicting canonicals, crawl failures, noindex directives, or whole URL groups that lack internal links. Compare old and new URL data rather than relying on the Change of Address status alone.
What does “Crawled - currently not indexed” mean after moving to a new domain?
It means Google crawled the new URL but has not currently included it in the index. It is not a precise diagnosis of poor quality. Check whether Google selected another canonical, whether the page is substantially duplicative, thin, poorly linked, empty when rendered, or inconsistent with the old-to-new redirect mapping. Compare excluded pages with similar pages that were indexed.
Should the old domain remain indexed after a site migration, and for how long?
The old domain should remain online to serve permanent one-to-one redirects for at least a year where possible, and ideally as long as relevant old links and bookmarks exist. Old URLs can remain visible temporarily while Google processes the move. They should not remain as a duplicate, self-canonicalized copy of the site; redirects should consistently point users and crawlers to the new domain.
How should Google Search Console be used after a domain migration?
Verify both properties, submit Change of Address from the old domain after redirects are live, submit the new canonical-only sitemap, and monitor Page indexing, Crawl stats, Performance, Manual Actions and Security Issues. Use URL Inspection for individual priority URLs and their old equivalents. Its live test validates current accessibility, while the indexed report shows what Google observed during its most recent crawl.
Sources
- https://www.reddit.com/r/bigseo/comments/1wzqlwm/domain_migration_from_old_highranking_site_4/
- 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/12482179?hl=en
- https://support.google.com/webmasters/answer/7440203?hl=en-IN
- https://developers.google.com/search/docs/crawling-indexing/301-redirects
- https://developers.google.com/search/docs/crawling-indexing/links-crawlable