Why index checks disagree

The short version

There is no single public "index lookup" for arbitrary URLs. Every method approximates Google's index from a different angle, so disagreement is expected. Knowing what each method actually reads is the way to decide which answer to trust.

What each method looks at

Google's Search Console help pages describe URL Inspection as showing information about Google's indexed version of a page, and they note that the live test only shows whether a page could be indexed, not that it is. The Page indexing report is built from the URLs Google knows on a property and is not refreshed in real time, so a page indexed today may still show under a "not indexed" reason for a while.

The common disagreements

Inspection says indexed, search finds nothing

The page is in the index but does not appear for the query used. This is the usual source of checker false negatives. Try other query forms, or wait and repeat. Index checker accuracy explains why search-based tools lean this way.

Search finds it, Inspection says not indexed

Less common. Usually the search result is a different URL (a canonical, a www variant, a version with parameters) that the tool or the person counted as a match. Compare the returned URL to the checked one character by character.

Report says not indexed, Inspection says indexed

Timing. The report lags behind individual inspections, so the newer reading wins.

Two checkers disagree

Different query forms, locations and match rules. A checker that requires an exact match on the returned URL will report more misses than one that accepts any result from the same path. Neither is wrong; they have chosen different trade-offs between false negatives and false positives.

Yesterday indexed, today not

Results move between datacenters, and Google sometimes drops pages it indexed earlier. A single flip is often noise; a URL that stays missing across several checks is more likely a real drop.

Canonical choice

When a URL has duplicates, Google picks one to index. Its documentation describes rel="canonical", redirects and sitemap inclusion as signals of differing strength rather than directives. If Google picks a different canonical than expected, methods that look up the exact URL will disagree with methods that look up content. Canonicalization and index checks goes through how to detect this.

How to resolve a disagreement

1. Normalize the URL: protocol, host, trailing slash, parameters.
2. If you own the site, run URL Inspection and check the "Google-selected canonical".
3. If you do not, run the full set of query forms in the single-URL method.
4. Re-check after a day or two before concluding anything.
5. For lists, choose one checker as the reference and stick with it, as recommended in bulk index checking.

The IndexChex checker counts a URL as indexed if any of four query forms returns it, which reduces disagreement caused by query choice but cannot remove differences caused by timing or canonicalization.