Crawling and indexing are two different stages of the same pipeline. Crawling means a bot fetched your URL. Indexing means the search engine evaluated the page and decided to store it in a searchable database. A page can be crawled many times and still never be indexed. Re-crawling won't change that unless the page itself changes.
Think of it this way: crawling is transport, indexing is admission, and serving is the window display. This article separates those three stages, shows the concrete signals that move a page from one stage to the next, and gives you a checklist you can run on a small site this week.
How to Tell Crawling, Indexing and Serving Apart
Before diagnosing anything, fix the vocabulary. A crawler is an automated program that requests a URL and reads the response. An index is a stored, searchable record of a page's content and signals. Serving is the moment a search engine or answer engine returns a result that links to or quotes the page.
These three stages fail independently. A page can return HTTP 200 to every crawler and still be absent from the index. A page can sit in the index and never be served for a query. A page can be served for a brand query while being absent for the topic it targets.
The table below shows what each stage actually requires.
| Stage | What happens | Typical failure | Where you see it |
|---|---|---|---|
| Crawl | Bot requests URL, reads HTML, fetches linked resources | Blocked by robots.txt, 4xx/5xx, timeouts, no internal links | Server logs; Search Console "Crawl stats" |
| Render | Browser engine executes JavaScript and builds the final DOM | Content injected client-side and never reached | URL Inspection "View crawled page"; rendered HTML |
| Index | Content parsed, canonical resolved, quality evaluated, stored | Duplicate, thin, or near-identical to existing pages | Page Indexing report; URL Inspection |
| Serve | Page ranked and returned for a query or AI answer | Relevant query never triggers it; snippet not selected | Search results; AI answer citations |
Google documents crawl, render and index as separate steps in its "How Search works" documentation. The render step matters because content added by JavaScript after the initial HTML response is not always seen on the first pass. If your body copy only exists after a client-side fetch, treat rendering as a separate check.
How to Read a 'Crawled - Currently Not Indexed' Status
This status in Google Search Console's Page Indexing report means exactly what it says. Google fetched the URL. Google did not add it to the index. It is not an error, and it is not a penalty.
Google's own documentation on page indexing statuses describes "Crawled - currently not indexed" as a page Google has visited but chose not to index. The word "currently" is doing real work: the decision can change after the page changes or after Google re-evaluates the site.
The most common misconception is that re-submitting the URL or requesting indexing again will fix it. Requesting indexing asks Google to look again. If nothing about the page changed, the answer is likely to be the same. The fix is almost always on the page or around it, not in the request.
There is no public figure for how long a crawled-but-unindexed page stays in that state. Google does not publish a timeline. A practical rule is to wait two to four weeks after a content change before re-checking, because re-checking sooner tells you nothing new.
How to Diagnose the Real Cause on a Small Site
Small sites have a specific problem: few pages, so each page carries more weight. On a large site, one thin page is noise. On a 30-page site, one thin page is a third of your topic coverage. Google's guidance on helpful content is that pages should provide substantial value relative to other pages on the same topic in the results.
Work through these checks in order. Stop at the first one that fails.
- Is the page reachable? Check robots.txt, meta robots, and X-Robots-Tag. A noindex tag or a robots.txt disallow will keep a page out of the index even after a successful crawl.
- Is the content in the initial HTML? View the raw source, not the rendered page. If the body text is not there, rendering is a suspect.
- Is the canonical correct? A canonical pointing elsewhere tells Google to index the other URL instead. Google's documentation on canonicalization explains that conflicting signals slow down indexing.
- Is the page a near-duplicate? Compare against your other pages and against the pages already ranking. If two of your pages answer the same query with the same words, one will be treated as the weaker copy.
- Is the page thin? Count the words, examples and specific facts. A page with 200 words and no specifics is a candidate for "not indexed".
- Does the page earn links? Internal links from your own pages help discovery. External links are a signal of value but are not required.
For a small site, the near-duplicate check matters most. A common pattern is a service page and a location page that share 80% of their text. Google's duplicate content guidance does not penalize this, but it does mean one page may be indexed and the other left out.
How to Use Logs and URL Inspection to Confirm the Stage
Server logs show the crawl stage with certainty. Search Console shows the index stage with a delay. Use both.
In your raw access log, filter for the Googlebot user agent and the URL. A 200 response with a full byte count means the page was fetched. A 304 means it was not re-fetched because it had not changed. A 5xx means the crawl failed. Cloudflare's documentation on verified bots explains how to confirm a request is really from Googlebot, not a spoof.
In Search Console, open URL Inspection for the specific URL. It shows the last crawl date, the crawl status, the indexed status, and a rendered screenshot. If the crawl date is recent and the status is "Crawled - currently not indexed", you have confirmed the stage. If the crawl date is old, the problem is discovery or crawl, not indexing.
One caveat for AI answer engines: OpenAI, Anthropic and Perplexity publish their own crawler user agents in their documentation. A page can be crawled by GPTBot and not cited by ChatGPT, because retrieval and citation are separate decisions from training crawl. Treat AI citation as a serving-stage question, not an indexing-stage one.
How to Fix a Page That Is Crawled but Not Indexed
There is no single fix. There are four fixes that match the four common causes.
Fix 1: Make the page different from your other pages. Rewrite the sections that overlap. Add specific examples, named sources, a table, or a checklist that no other page on your site has. The goal is that a reader comparing two of your pages can tell why both exist.
Fix 2: Add substance. A page that answers a question in 150 words is a candidate for consolidation, not expansion. Either merge it into a stronger page or give it enough specific content to stand alone. Google's helpful content guidance frames this as people-first value, not word count.
Fix 3: Remove conflicting signals. Check canonical, hreflang, robots meta and sitemap entries. If your sitemap lists a URL that your canonical points away from, fix one of them. Google's sitemap documentation states that sitemap URLs should be canonical.
Fix 4: Improve discovery. Link to the page from your strongest pages with descriptive anchor text. A page that only appears in the sitemap and nowhere in navigation is harder for a crawler to prioritize. For a small site, three or four internal links from relevant pages is usually enough.
After the change, request indexing once. Then wait. Re-requesting indexing daily does not speed up evaluation and can dilute the signal that something changed.
How to Audit Thin and Duplicate Pages on a Small Site
Run this audit once a quarter. It takes about an hour on a site under 100 pages.
- Export all URLs from your sitemap and from Search Console's Page Indexing report.
- Group URLs by topic. Any group with more than three near-identical pages is a candidate for consolidation.
- For each URL, record word count, number of specific examples, and number of internal links pointing to it.
- Flag pages under 300 words with fewer than two internal links. These are the most likely to sit in "Crawled - currently not indexed".
- Decide per page: expand, merge, or remove. A 301 redirect from a removed page to the merged page is cleaner than leaving both live.
Use the same audit for AI answer engines. A page that is thin for Google is also thin for retrieval. The retrieval step in an AI answer pipeline pulls passages, not whole pages, so a page with one strong paragraph and no supporting detail is a weak retrieval candidate.
How to Decide When to Wait and When to Act
Waiting is correct when the page is new (under four weeks), when the crawl date is recent, and when the page is already distinct from your other pages. New pages on small sites often take several weeks to move from crawled to indexed. Google does not publish a fixed timeline for this.
Acting is correct when the page has been crawled and unindexed for more than four weeks, when the page overlaps another page on your site, or when the page has fewer than 300 words and no specific facts.
A useful test: open the page and ask whether a stranger would learn something from it that they could not learn from your other pages. If the answer is no, the page is a consolidation candidate, not an indexing problem.
Key Facts at a Glance
- Crawling fetches a URL; indexing stores an evaluated record; serving returns it for a query. They fail independently.
- "Crawled - currently not indexed" in Google Search Console means Google fetched the page and chose not to index it. It is not a penalty.
- Requesting indexing again does not fix the status if the page itself has not changed.
- On a small site, near-duplicate pages and thin pages are the most common causes.
- Server logs confirm the crawl stage; URL Inspection confirms the index stage.
- AI answer engines run retrieval and citation decisions separately from training crawl, so a crawled page is not automatically a citable one.
FAQ
What does "crawled - currently not indexed" actually mean?
It means Googlebot fetched the URL and Google decided not to add it to the index at that time. The page is not blocked and not penalized. The decision can change after the page changes or after Google re-evaluates the site.
How long does it take Google to index a new page?
Google does not publish a fixed timeline. On small sites, new pages often take several weeks to move from crawled to indexed. The practical check is to wait two to four weeks after a content change before re-inspecting.
Does crawl budget matter for a small site?
Rarely. Crawl budget matters when a site has many URLs and limited crawl capacity. A site under a few hundred pages is unlikely to hit a crawl budget limit. Google's crawl budget documentation frames it as a concern for large or frequently changing sites.
Will re-submitting the URL in Search Console fix it?
Only if something about the page changed. Requesting indexing asks Google to look again. If the content, canonical or internal links are unchanged, the outcome is likely the same.
Does a page need to be indexed to be cited by an AI answer engine?
Not necessarily. AI answer engines use their own retrieval systems, which may fetch pages directly. A page that is crawled but not indexed by Google can still be retrieved and cited by an AI engine. Check the engine's own crawler documentation for its user agent and behavior.