"Crawled - currently not indexed" has no fix button. Here's the decision Google actually asks you to make.

Search Console’s Why pages aren’t indexed table gives every reason a definition — what happened and, usually, what to do. Here is Crawled - currently not indexed’s, complete, from Google’s help page for the Page indexing report:

Google’s help page — the definition, in full

The page was crawled by Google but not indexed. It may or may not be indexed in the future; no need to resubmit this URL for crawling.

Two sentences. Two trade words, translated once: crawled — Google’s reading program visited and read the page; that part worked. Indexed — taken into Google’s index, the store of pages it reaches into to answer a search. Translated whole: your page was read, and then passed over.

Now read it again for what is not in it. No cause. No timescale beyond “may or may not”. No fix — the only instruction is negative: no need to send the page in again. The reasons around it each name what happened — an instruction found on the page (that one has its own article), a file asking Google to stay out, an address that answered “not found”. This one names nothing, points nowhere — and nothing in Search Console will show why a given page is in it.

Three worries arrive with this row. Is it serious? Usually not — Google’s own FAQ answers with one word, below. Did you break something? No — this row points at no switch, no file, no line of yours. The report itself says so: its tables carry a Source column — the help page: “In general, you can fix only issues where the source is listed as ‘Website’” — and on my own site this row’s Source reads Google systems: the event filed on Google’s side of the fence. A Product Expert made the same point in a 2022 thread (a forum title Google pins on trusted volunteers, not employment): anything that is your fault “would be reported more explicitly”, whereas this row: “There may be many reasons why a page is not indexed. Or may be no specific reasons.” Does the size of the number matter? Not as a scale of harm. What matters is which pages, never how many.

First, make sure it’s this row and not its neighbour

One row away sits a near-twin, Discovered - currently not indexed: one word different, completely different meaning. Google’s help page defines the neighbour as “The page was found by Google, but not crawled yet”, and notes: “This is why the last crawl date is empty on the report.”

Discovered means Google knows the address and has not read it — a queue, an implied not yet. Crawled means Google read it and declined — a verdict, no reason attached. That last sentence hands you the check: open either row — its Examples table has a column headed Last crawled. A date there means you are in this article; an empty cell means the waiting state. In July my own report held both — 102 pages under Crawled, 3 under Discovered. Don’t assume you have only one; read the row’s name first.

How serious is this? Google has already answered

The help page’s own FAQ asks this article’s question in as many words — “Is it OK if a page isn’t indexed?” — and answers it in one: “Absolutely.” The rest of that answer names kinds of pages with named causes; the row you are holding is the one the list leaves out, which is why the deciding, further down, falls to you. The comfort is repeated in the same page’s list of expectations to drop — “Don’t expect every URL on your site to be indexed” — and its guide for non-specialists even braces you, word for word: “If you don’t understand indexing, this report will confuse or frustrate you — trust us.”

So the count is not the width of a wound. A site usually answers at more addresses than it has pages, because platforms mint addresses on their own: one for every month of a blog’s archive, one for every tag, one for every way a shop’s products can be sorted, and an old design leaves its addresses behind. Rows like this collect that overflow. What the count cannot tell you is the only thing that matters: whether any page you would send a customer to is in there. That check is Google’s own and takes minutes.

The five-minute check Google actually prescribes

For small sites, the same help page is direct: “If your site has fewer than 500 pages, you probably don’t need to use this report”, and prescribes ordinary Google searches instead. The searches use the site: prefix — type site: followed by your domain (the short name of your site, no www., no https://) into an ordinary Google search box, and Google lists what it holds from your site. And for one specific page, Google’s URL Inspection documentation names a test stronger than the tool itself: “For a definitive test to see if your URL is on Google, search for your URL on Google.”

Paste the page’s full address into Google’s search box. If it comes back — anywhere among the results, not necessarily alone or first — it is listed, whatever any report says. If it doesn’t, it isn’t.

And if the site: search returns nothing at all — not one page — you are still in this article’s story, at its widest case. For a site that is weeks old, Google’s own indexing FAQ offers the boring explanation first: the site may be “just too new”. For an older site, it is the fourth ending below, met early.

Run those two searches for the pages you actually care about — home, services, products, the pages you’d print on a business card. All there? This row holds only overflow — Google’s FAQ has already answered it: absolutely fine. If a page you care about is missing, keep reading — the next part is what other guides soften.

The button that isn’t there

Every other article in this series ends with proof — run the live test, watch the answer arrive. Here, Google documents that its own tools cannot test this state. The URL Inspection documentation lists what the live test cannot check:

Google’s URL Inspection documentation

Here are indexing issues from the Page Indexing report that can’t be tested in the live test: […] Crawled - currently not indexed Discovered - currently not indexed

And when the same page asks whether any test can settle indexing, the answer is flat: “There is no definitive test that can guarantee whether your page will be included in the Google index.”

What Google offers instead, on the same inspection page: “Read the details in the Page indexing section to learn why the page wasn’t indexed, and decide whether this is something that you need to fix.” For this row, “the details” are the two sentences this article opened with — which leaves the last clause as the whole instruction: not a test you run on the page, but a decision you make about it. Does this page need to be listed at all?

This also settles the diagnosis half of whether to pay someone: nobody you can hire can read a reason Google does not show — if someone says they know why your particular page is out, ask what they read it from; the tool that could prove them right does not exist, and Google’s own page just said so. What a good specialist can fairly be paid for is further down.

What happened when I pressed the button anyway

The row’s screen in Search Console offers one more button: Validate fix. At the end of July, with the row on one of my sites holding 102 pages, I pressed it.

The verdict came back Failed, and the details strip keeps its own dates: started 7/31/26, failed 8/5/26 — five days, start to verdict. Nothing on the site had got worse. The button had asked Google to confirm a repair, and nothing had ever been reported broken: validation re-reads pages for the absence of a named issue, and here the “issue” was Google’s own decision not to list them. My pages came back still unlisted — which is all the red verdict has ever meant on this row, and which the report’s latest update, dated 8/21/26, still prints as a tally: PENDING 80, FAILED 5. (What “Failed” means in general has its own article.)

The Product Expert quoted earlier described exactly this, four years earlier, in the same 2022 thread, spelling his: “The ‘validation’ fails, because you can’t ‘fix’ it. Google wont tell you why it didnt index it in the first place.”

Five addresses from my own row

For most of the summer my notes held counts, dates and a verdict, never an address; the failed validation ended that. Its details page keeps an Examples table of the pages it re-read and failed — the five from the tally above. For once, the example pages are mine — all five, as the report lists them:

• https://kovalweb.com/uk/privacy-policy/ — Last crawled Aug 22, 2026
• https://kovalweb.com/what-is-ssl-certificate/ — Last crawled Aug 11, 2026
• https://kovalweb.com/what-is-seo/ — Last crawled Aug 7, 2026
• https://kovalweb.com/how-to-create-online-store/ — Last crawled Aug 7, 2026
• https://kovalweb.com/ru/how-to-create-online-store/ — Last crawled Aug 1, 2026

I opened them at the end of August. All five load, and all five answer 200 — the server code for a healthy reply. None is empty, none is an error page, none is a stub: counting only readable text, they run from roughly nine hundred words to nineteen hundred. Google has read every one — the dates say when — and, as the report last stood, every one is out.

Handle those word counts with care: they close a door, nothing more. The first guess everyone reaches for on this row is the pages must be thin, and these five are visibly not — that is all the numbers may say, because no grade exists: Google’s own list of questions for people who write content asks “Are you writing to a particular word count because you’ve heard or read that Google has a preferred word count?” — and answers itself: “(No, we don’t.)” What the five addresses show is narrower: the obvious explanation is visibly absent, and Google does not say what the actual one is. (The last two are the same article in two languages, and both sit in the row — but nothing in my report says that is why.)

The advice you will meet, and what stands behind it

This row hands its advisers nothing — no cause, no test — and so attracts more confident advice than any other warning in this series. Most people offering it are trying to help; some answer Google’s forum for free, year after year. Even the guide most often linked as the semi-official answer was written by one of those volunteers, not by Google — and its central sentence is at least candid about the limits: “There’s no magic bullet to fix this, as Google’s judgment on what to index can be subjective.” Each circulating remedy, held against the record:

“Press Request indexing on each page.” Contradicted three times over in Google’s own pages: the URL Inspection page says “Submitting a request does not guarantee that the page will appear in the Google Index”; its page about asking for a re-read says inclusion is not guaranteed “instantly or even at all”; and the row’s own definition closed with “no need to resubmit this URL for crawling”. The fairest thing the record shows: one owner in Google’s forum reported that pages he resubmitted “are indexed 100% of the time” — and the volunteer expert answering called that “a little, temporary, shove over the line”. A lever that sometimes moves one page for a while is why this advice survives. It is not a fix.

“Add internal links; get backlinks; raise your domain authority.” No Google source ties links — your own pages’ or other sites’ — to leaving this row; the nearest thing on record is about the site as a whole, in the next section. And “domain authority” is not a Google concept; Google publishes no such number.

“Submit a sitemap — or a temporary one.” A sitemap is the address list a site publishes for search engines; Google’s own documentation says it “doesn’t guarantee that all the items in your sitemap will be crawled and indexed”. A page in this row was already found and read — discovery was never its problem. The “temporary sitemap” trick has no Google source I could find at all.

“Fix your crawl budget.” Contradicted, not merely unsourced. “Crawl budget” is the trade name for how much reading time Google gives a site, and Google’s guide to it is written for sites with a million-plus pages; the only row it names is the neighbour, Discovered - currently not indexed; this article’s row appears nowhere on that page.

“Just wait — it fixes itself” and “it can last forever.” Two clocks, and neither has a source. Google’s entire forward-looking statement in writing is the definition’s “may or may not”; the only spoken addition, below, is “once we’re happy”. And “forever” traces to a remark about the Discovered row, not this one. Nobody has the clock.

The one direction with something behind it

Across fifteen official office-hours transcripts, the Search Central blog and the documentation pages this article draws on, the row’s name surfaces exactly once — on the list, quoted above, of what the live test cannot check — and nothing in any of it says what the row means. The only Google words that go past the two-sentence definition are in a July 2026 episode of its podcast, Search Off the Record — Google publishes the episode’s transcript.

In it, Martin Splitt of Google asked John Mueller whether this row is “often or only sometimes a sign of a quality issue”: “Sometimes. So it’s definitely the case if our systems are seriously worried about the quality of a website, that they will reduce the number of pages that they index.” Then, whose clock this runs on (the same transcript): “And once we’re happy — we looked at it — once we’re happy, we will take another look and see if we can index it.” That is everything Google has published — and notice what it is not: a list of steps. It is also about the site more than the page: asked in June 2021 about pages sitting in exactly this state, Mueller replied on Twitter that “it’s normal that we don’t index all pages on all websites”, that it is “not an issue with ‘that page’, it’s more site-wide”, and that making sure the site is “of the highest quality possible is essentially the direction”.

What does “work on the quality” mean when nobody will mark it? Google’s most concrete offer is a page of questions to ask about your own content — self-assessment, not a checker. One of them is this row’s economics turned back on you: “Does the content provide substantial value when compared to other pages in search results?” Search for what your page is about; read what Google lists instead. A page that would not earn its place on that screen is the closest thing to a diagnosis this row allows — and it points to real work on the page and the site around it, not a setting. A page that plainly would earn it means you have met the limit of what anyone outside Google can know — what remains is the decision above, and a clock you don’t control.

And if that work is not yours to do — someone else built the site — the work itself can honestly be bought: not the diagnosis (nobody has that), but the craft. The comparison you just ran is the brief to hand over: here are the pages I most need found — search what each is about, read what Google lists, and make my page one that earns its place on that screen. Check the result yourself by running the same search again when the work is done.

Where that leaves you

Four readers reach this row. Each gets an ending.

• Nothing to do The listed pages are ones you would never send a customer to. Month archives, tag pages, sort orders, addresses an old design left behind. Google’s FAQ answered this case in one word; the searches above confirm your real pages are safe. Close the report — that is the fix.
• Ten minutes A small site, and you only want to know nothing precious is lost. Run the two searches on your key pages. If they come back, Google’s own guidance says you probably didn’t need this report. You checked the shelf instead of the paperwork. Done.
• Real, and slow A page you genuinely care about is out. The address search proved it. No button moves it — the direction is the previous section’s: make the page, and the site around it, something that wins its own comparison. The re-evaluation, when it comes, will be on Google’s clock and unannounced — so re-run the address search from time to time. Watch the search, not the row.
• The hard one Most of the site sits in the row. When Google’s systems doubt a site as a whole, they index less of it, and the direction is site-wide — the site entire, not single pages. A volunteer expert told a stuck owner this July, in the plainest words in this record: “It depends. I don’t think there is an exact timeframe. Also, I have not seen many sites recovering from the ‘Crawled - currently not indexed’ issue, so it might take some time.” Weigh that sentence: it is one person’s memory, not a count — and a help forum only ever hears from the stuck. In the eighteen forum threads read for this article, no owner ever reported this row’s count moving. On my site it moved — 111 pages on 9 July, 102 at the end of July, 90 on 17 August, 85 at the chart’s edge on 21 August — twenty-six addresses out of the row in six weeks, while pages indexed held roughly level at 93. I was working on that site the whole time, so the fall is no testimony for waiting, and it promises nobody a recovery — but it retires the bleakest reading of the volunteer’s sentence: the number is not frozen. It moves. And “once we’re happy, we will take another look” is not indifference — it is Google’s own person saying re-evaluation is standing policy. The door is not closed; it is unattended. The meanwhile is the same decision, made site-wide: decide which of those pages should be listed at all — “Don’t expect every URL on your site to be indexed” is Google’s sentence, not mine — and put real work into the ones that should.

If you already pressed Validate fix: nothing is damaged — mine failed too, and the section above says why. Read the red strip as the definition holding, not a second verdict.

The row’s own name turns out to have been carefully written. Not “rejected”. Not “excluded”. Currently not indexed — Google’s word, holding the door ajar. Beneath it, the one instruction that survived every test this article ran: decide whether this is something that you need to fix. That decision is the button — the only one this row ever shipped with. Google left it out of the interface, and left it with you.

Written by Oleh Koval at Specia1ne, who reads these reports for a living.