Technical SEO · 2026-11-18
Discovered – currently not indexed: what it means and how to fix it
Google found the URL and has not crawled it yet. What that Search Console status means, how it differs from Crawled, and which checks show what to fix.
In Google Search Console, "Discovered – currently not indexed" means Google knows the URL and has not fetched it. The Page indexing report says: "The page was found by Google, but not crawled yet. Typically, Google wanted to crawl the URL but this was expected to overload the site; therefore Google rescheduled the crawl. This is why the last crawl date is empty on the report."
Googlebot has not read the page, so nothing on it has been accepted or rejected. For a URL you care about, the work is to make a crawl possible and worth doing: the server has to answer, the URL has to be one Google can justify fetching, and it has to be findable from a link or a sitemap.
Discovered is not Crawled
Both rows live under Why pages aren't indexed. They are different events.
- Discovered – currently not indexed. Found, not crawled. The last crawl date is empty. The help text's usual reason is a crawl postponed so it would not overload the site.
- Crawled – currently not indexed. Fetched, not indexed. Google says it does not index every page on the web, so the message is common and does not necessarily mean something is broken. The practical read is in Crawled – currently not indexed: what it means and how to fix it.
Indexing is a later decision. The crawl budget guide says that for Google Search, not every crawled page is indexed. A fetch can put the URL in the index or move it to "Crawled – currently not indexed." If an important page lands there, server tweaks will not change a page Google has already read.
On 10 November 2026, John Mueller was asked on Bluesky what to do when a large share of a site stays in one of these two statuses. He said that usually means Google's quality systems are unconvinced about the site, or convinced in a way you do not want, and that making the site smaller does not significantly raise overall quality. He called quality and crawl budget distinct. A tiny indexed share, he wrote, suggests either a severe technical issue — "crawl budget," which you would definitely notice at that scale — or that Google is not convinced about the site. The posts are here and here, and the short account is Mueller: shrinking a site will not fix a mass not-indexed problem.
An empty last crawl date marks that postponed fetch, which is separate from the quality comment. The same week he said noindex is fine when you do not want a URL indexed — the right tool for keeping a URL out, not for raising the quality of the URLs you want in (post).
What "overload" refers to
That crawl budget guide is for large or fast-changing sites. One audience it names is a property with a large portion of its URLs in this status. The scale examples are rough, not cut-offs: 1 million or more unique pages changing about weekly, or 10,000 or more changing daily. If you do not have a large set of pages changing rapidly, or if pages are crawled the day they are published, Google says you can skip the guide. An updated sitemap and a regular look at the Page indexing report are enough.
A site, in that guide, is one hostname with its own budget. https://www.example.com/ and https://code.example.com/ are separate sites. The budget is set by a crawl capacity limit, also called hostload, and by crawl demand. Capacity rises when responses stay consistent and response times, including latency and time to first byte, hold steady or improve. It falls on slower responses, HTTP 5xx, or a rate-limit such as HTTP 429. Demand for Googlebot depends on size, how often pages change, quality, and relevance. Low demand means less crawling even when the capacity limit still has room.
A long discovered list on a calm server is an inventory and demand problem. A slowing or erroring server is a capacity problem. The guide's two levers match that: more server resources when URL Inspection shows "Hostload exceeded," and the Search factors it names for how crawling is allocated — popularity, user value, uniqueness, and serving capacity.
Check it in Search Console
-
Open the Page indexing report and click Discovered – currently not indexed under Why pages aren't indexed. Read the URLs and the history, not only the count. Not indexed can be legitimate: Google says not to expect every URL indexed, only the canonical of each important page. Check Source. You can generally fix issues marked "Website."
-
Look for a pattern: one folder, one template, a parameter, or one publishing burst. A few important HTML URLs are a different problem from a tail of filtered addresses.
-
Filter with the sitemap dropdown: all known pages, all submitted pages, unsubmitted pages only, or one sitemap. A URL in a sitemap submitted in the Sitemaps report, or listed in robots.txt, counts as submitted even if a link also revealed it. If your important sitemap is indexed and the discovered pile is unsubmitted combinations, the report is describing the tail.
-
Inspect URLs you want indexed with the URL Inspection tool. The stored view is not a live test. Under Page indexing, read Discovery, Sitemaps, Referring page, Last crawl, Crawl allowed?, and Page fetch. No referring sitemap means Google could not identify a sitemap for that URL. A blank Referring page does not prove there is no link: the tool may simply lack that information. An empty Last crawl is what this status predicts. "URL is unknown to Google" means Google has not seen the URL at all.
-
Run Live test on a public URL that needs no login. Google says that test is fine even when the page is not indexed. A response that returns the real page shows a fetch is possible now. It does not index the URL. "Hostload exceeded" is the capacity example in the crawl budget guide.
-
Open Settings, then Crawl stats. The Crawl Stats report is only on a root-level property, a domain or a URL-prefix at the host root, and Google says a site under a thousand pages should not need it. Host status covers 90 days: no significant issues, an older issue in that window, or an issue in the last week. Average response time counts every fetched resource, so one slow file type can move it. The response table shows 5xx, timeouts, DNS failures, and an unavailable robots.txt. A rising server-error rate makes Googlebot throttle, and a robots.txt that stays unavailable for a day pauses crawling.
-
Confirm those URLs are linked from known pages and listed in a submitted sitemap. A sitemap names the files you consider important. It does not guarantee a crawl or indexing, and linked pages are usually discovered anyway. It helps more on a large site, a new site with few outside links, or a site that orphans URLs. Google's rough case for skipping one is about 500 pages or fewer that you want in Search, already linked throughout.
The fix has to match the cause
A slow server, or errors
If host status flagged the last week, response times lengthened, or the table shows 5xx or timeouts, fix that before asking for more crawling. Extra capacity is the documented answer to "Hostload exceeded" when the host is actually the limit. A server with room left does not gain fetches from a bigger machine.
Read the status code, not only the page. A 200 that says "back soon" can be indexed with that text, and URLs sharing that body can be grouped as duplicates. John Mueller has described both: the error text clears when the real page returns, and the duplicate grouping lingers because duplicates are crawled less often. For 5xx and 429, crawlers slow down, 5xx content is ignored, and indexed URLs are kept, then eventually dropped. Persistent server errors are removed. After a return to successful 2xx responses, the crawl rate rises gradually. For an outage of a day or two, the documented signal is 503. See a 200 "back soon" page can cost indexed URLs.
Too many low-value URLs
Google tries to crawl URLs it knows. Filters, sorts, tracking parameters, and site search can multiply one page into a large set. The Page indexing report notes that many URLs with the same content can time out, or the server can return an overloaded status and ask Googlebot to slow down. Keep parameter lists short.
Consolidate duplicates. Block URLs you never want crawled with robots.txt. Noindex does not save the fetch: Google still requests the URL, then drops it. Use 404 or 410 for pages that are gone. A blocked URL stays in the queue much longer than a gone one.
Google will not move that freed budget onto other pages unless the site is already at its crawl capacity limit. If the host has room and an important URL is still only discovered, look at links, the sitemap, and demand. Cutting the tail does not raise quality. Mueller said making a site smaller does not significantly raise it.
Pages that are hard to find
Google needs a link from a known page, or a sitemap. On a new site, request indexing of the homepage so crawling has a start. Link important URLs from pages already being crawled, list them in the sitemap, and include lastmod when the content changes. Then check URL Inspection for a referring sitemap you submitted and a referring page Google possibly followed.
Already crawled, still not indexed
A last crawl date means this status is the wrong label. Follow the Crawled guide. On 9 November 2026, asked about failed indexing requests on a fairly new site, Mueller said to focus on normal crawling and indexing, and that a burst of new content takes time to understand — quality over quantity (post). Wait on a healthy new site. It is not a promise that every URL gets indexed.
When to wait, and when to validate
Wait when the important URLs are linked and submitted, a live test returns them, robots.txt is not blocking them, and host status is calm on a site large enough for Crawl stats. New pages can take a few days to be indexed, and requesting indexing can shorten that lag for a few important URLs. This status is a crawl Google delayed so it would not overload the host, so do not queue the whole discovered list. URLs you never wanted crawled can stay unfetched.
Validate fix is optional, and only after a real change. Google updates the count on ordinary crawls too. To speed a request up, submit a sitemap of only your most important pages, filter the report to it, then validate. A subset can finish faster, and with the filter on, validation covers only that sitemap's URLs at the moment you click.
Click Validate fix once. Do not click it again until that attempt has succeeded or failed. It typically takes up to about two weeks, and sometimes much longer. Google checks a few pages first and stops if the issue is still there. Otherwise it queues the known URLs for this issue, not the whole site. If validation fails, fix the URL that failed, confirm anything still pending, and start a new one. URLs already marked passed or other are not rechecked.
What "validation passed" means
Passed means every known instance of the issue is gone, or the URL is no longer available, and only after you click Validate fix. If the rows clear on a normal crawl and you never clicked, the state is N/A. Leaving this row is not the same as being indexed. The URL may be indexed, it may be Crawled – currently not indexed, or it may be removed, noindexed, or behind a login, which can count as Other. Inspect the URL and read the new reason.
FAQ
What does "Discovered – currently not indexed" mean in Search Console?
Google found the URL and has not crawled it. Typically it wanted to crawl, expected that to overload the site, and rescheduled, which is why the last crawl date is empty.
How is it different from "Crawled – currently not indexed"?
Crawled means the fetch happened and the page was not indexed. Discovered means the fetch has not happened. An empty last crawl date is the fetch-side clue.
Why would Google discover a page and not crawl it?
The status text points at a crawl rescheduled to avoid overload. Capacity also drops on slow responses, 5xx, and 429, and demand can be low when many known URLs are duplicates. A sitemap or internal link puts a URL on the known list. Neither forces a crawl or an index.
How do I fix "Discovered – currently not indexed"?
Separate URLs you want indexed from URLs you do not want crawled. Inspect the important ones, fix a real server problem, block junk in robots.txt, and link plus submit the pages that matter. Then wait, or validate that subset once. Deleting pages to shrink the report does not raise quality.
What does "validation passed" mean for this status?
The known instances of this issue are gone, or those URLs are no longer available to Google. Inspect each important URL for its new status. Passed is not a ranking, and it is not a reason to click Validate fix again.
How long does the fix take to show?
New content can take a few days to be indexed. Validation typically takes up to about two weeks, and sometimes much longer. Google publishes no day count on which this status expires. If an important URL still has an empty last crawl, inspect it again instead of starting a second validation.
Checklist
- Open Page indexing, then Why pages aren't indexed, then Discovered – currently not indexed, and read the URLs.
- Filter to the sitemap of pages you want indexed.
- Inspect a sample. Last crawl empty, crawl allowed, live test returning the real page.
- On a large site: Settings, then Crawl stats. Host status, average response time, 5xx, timeouts, robots.txt.
- Link important URLs from crawled pages and list them in a sitemap. Block never-crawl URLs in robots.txt. Save noindex for URLs you do not want indexed and will allow Google to fetch.
- Do not shrink the site or press Validate fix during a run that has not succeeded or failed.
- After a real fix, validate the important sitemap once, allow up to about two weeks or longer, then inspect again.
Query and click reviews only see a URL after a crawl. Until the last crawl date is filled in, there is nothing there to measure.