GSC: Crawled but not indexed

In the Search Off the Record podcast, Google's John Mueller and Martin Splitt examined the indexing report in Google Search Console in more detail. The discussion focused on why URLs can be found and crawled yet still not enter the index. Especially relevant for SEO teams is the status “Crawled – currently not indexed,” which often relates to quality issues – including low-quality AI-generated content.

What the indexing report really shows

Google Search Console separates several stages on the path from discovery to indexing. “Discovered” means Google knows a URL but has not crawled it yet or has not crawled it again. “Crawled – currently not indexed,” by contrast, shows that the bot retrieved the page but currently decides against indexing. This is not a technical crawl failure; it is an assessment of priority and presumed quality.

That distinction matters in daily work. Many teams read the status as a pure budget or server problem. Mueller and Splitt made clear that quality and trust signals can play a central role. If Google rates the content as weak, thin, or unhelpful, the page can be crawled and still be kept out of the index.

Quality issues as an indexing brake

In this context, quality is not an abstract score but the sum of usefulness, originality, clarity, and trustworthiness. Pages with thin text, product lists without added value, mass-generated text blocks, or missing topical depth risk not being released for searchers. The crawl only confirms reachability; indexing follows only when the content justifies the expected value.

Signals around E-E-A-T also matter: experience, expertise, authoritativeness, and trustworthiness. Without verifiable sources, clear authorship, or a recognizable editorial standard, the chance of stable indexing drops. Especially for topics with high information needs, Google expects substantive content rather than shallow summaries.

AI content and missing trust

A key point from the podcast: low-quality AI-generated content can cause pages not to be trusted and not to be indexed. The decisive factor is not the use of AI itself, but output quality. Texts that sound generic, mix facts, show no original research, or only cover search intent superficially raise the risk of “Crawled – currently not indexed.”

For editorial teams, that means AI can speed up drafts but does not replace editorial review. Facts must be verified, claims sharpened, and examples made concrete. Without human finishing, content can be crawlable and still be judged untrustworthy for the index.

Discovered versus Crawled Not Indexed

“Discovered” and “Crawled – currently not indexed” require different actions. For “Discovered,” internal linking, sitemap maintenance, crawl budget, and technical reachability matter. For “Crawled – currently not indexed,” the focus shifts to content, differentiation, and page quality. Teams that only check robots rules or server logs often miss the real cause.

• Discovered: the URL is known; crawling is pending or delayed.
• Crawled – currently not indexed: crawling occurred; indexing is deferred.
• Quality and trust can block indexing despite a successful crawl.
• Weakly produced AI content is a concrete risk example from the podcast.

In practice, systematic prioritization helps: first review URLs with strategic traffic potential, then clean up duplicates and thin content, and only then push for indexing again. Resubmitting in Search Console without content improvements rarely changes the quality assessment.

Actions for SEO and content teams

Teams that want to reduce “Crawled – currently not indexed” sustainably should make content quality measurable. That includes clear search intent, unique sections, current data, transparent sources, and a recognizable advantage over competitors. Pages that merely paraphrase existing rankings often remain in crawl limbo.

Technically, the basics still matter: clean canonicals, consistent internal links, sensible indexing controls, and avoiding soft-404 patterns. Those factors alone do not solve the quality problem. Mueller and Splitt suggest that Google tends to index cautiously when usefulness is uncertain rather than releasing weak pages.

Editorial control for AI-assisted texts

AI-assisted workflows need binding quality gates: topic briefings with target intent, mandatory source checks, editorial rewriting, and approval by subject-matter owners. Especially in scaling content programs, every URL should have a clear purpose. Pages without value for users or the brand unnecessarily burden crawling and indexing.

Closer collaboration between SEO and editorial teams also helps. Indexing reports should be linked regularly with content audits. That reveals whether certain templates, author profiles, or automated series systematically produce “currently not indexed.” Targeted fixes then replace blanket indexing requests.

Practical takeaway

The Search Off the Record comments reinforce a known but often underestimated logic: crawling is necessary; indexing is earned. The Search Console report makes that separation transparent. Teams that ignore quality issues – including weak AI content – risk that Google sees and fetches pages without serving them in search results.

SEO leads should treat the indexing report as a quality radar. Instead of single URLs, look for patterns: which content types stall, which clusters look thin, where expertise is missing. Those levers address “Crawled – currently not indexed.”

For teams scaling with AI: automation without editorial substance weakens indexing chances. Using AI for research and drafting, with a final version aligned to user needs, improves crawl and index outcomes.