Why a page is not indexed
A page can be missing from the index for three distinct reasons: it was not discovered, it could not be crawled, or it was crawled and then discarded. The engine publishes a state for every page, and that state says which of the three applies. Indexing is never guaranteed.
Why does “why is my page not indexed” not have one answer?
Because it has three, and they have nothing to do with one another. A page may not have been discovered, may have been discovered without being downloaded, or may have been downloaded and then discarded. All three produce the same symptom — the page does not appear — and call for three unrelated pieces of work.
That is why the diagnosis always starts by reading a state, never by editing a page. The engine publishes that state for every address it knows, and it names it.
The indexing report separates “discovered, currently not indexed” — the page has not been crawled yet — from “crawled, currently not indexed” — the page was downloaded and then discarded.
officialSource : Google Search Console, “Page indexing report”
Those two labels look alike and they describe two different worlds. The first says: we have not looked yet. The second says: we looked and we did not keep it.

What do you check first, and why in that order?
One — is the page reachable by the crawler? This check comes before every other, because a block makes everything that follows pointless. Three configurations produce a block: a disallow in the exclusion file, an error response from the server, and a requirement to authenticate. The indexing report names each of them separately, including redirect errors and “soft” not-found pages that answer as though they were valid.
Two — does the page carry a no-index instruction? This is the most frequent cause and the most embarrassing, because it is almost always unintentional: the instruction comes from the staging environment and the launch forgot to remove it.
The engine may index any address it finds, unless the page carries a no-index instruction.
officialSource : Google Search Console, “Page indexing report”
That sentence reads in both directions, and the second is the more useful one: if there is no no-index instruction, the address is indexable. In other words, the question is never “how do I allow indexing”, but “what is preventing it”.
Three — is the address being treated as a duplicate? A page can be perfectly reachable and still be absent because the engine attached its content to another address.
Among the published states is one for a duplicate page where the engine kept a canonical address different from the one the site declared.
officialSource : Google Search Console, “Page indexing report”
This case deserves careful reading: the page is not rejected, it is grouped. Its content is present in the index under another address. The job is therefore not to make it indexable, but to work out why two addresses carry content judged equivalent — the subject of the guide on canonicals.
Reading the indexing state before editing the page
hypothesis
- Effort
- low
- Time to effect
- immediate
- Risk
- none
- Durability
- structural
The trap that produces the opposite of what was wanted
It concerns the pages you want taken down, and it is explicitly documented:
If a page is blocked by the robots.txt file, the crawler will never see its no-index instruction, and the page can stay visible in the results.
officialSource : Google Search Central, “Block Search indexing with noindex”
The mechanism is clear once crawling and indexing are held apart: the instruction lives inside the page, so blocking access to the page blocks the reading of the instruction. The reflex of “block everything to be safe” produces exactly the result it was trying to avoid.
The practical consequence is easy to remember: to remove a page, you let the crawler in. To save crawling, you block — but then you give up any instruction carried by the page itself.
What do you do when the page is crawled and still not indexed?
That is the case left once the previous three have been ruled out, and it is the hardest to hear.
The engine does not guarantee that every page, everywhere, will enter its index.
officialSource : Google Search Console, “Page indexing report”
So there is no technical remedy, because there is no technical defect.
A page that was crawled and then not indexed is the only state that points back at the worth of the page itself, and it is also the one no technical setting corrects.
hypothesis
That “hypothesis” status is owned up to: the engine does not say why it discards one page in particular. What you can do with this state is therefore a reading, not a procedure. The useful question becomes: does this page bring something the pages already indexed on the same subject do not? If the honest answer is no, no setting will save it, and merging it with another page beats maintaining it.
What does not work, or no longer works
Page not indexed? Submit it three times in the inspection tool, add it to the sitemap, then send it links from social networks to force indexing.
There is also a family of services selling indexing as a product. They exploit the confusion between the three states: on a page that was merely undiscovered, any public link will produce the expected effect and the service will look as though it works. On a discarded page, nothing will happen, and the failure will be blamed on the site.
The method that remains is less saleable: read the state, fix what genuinely blocks, then accept that the last question is editorial. It has the advantage of asking you to believe nobody — every state is published, named, and checkable in a public document.
Sources
- Page indexing report — Google Search Consoleaccessed on
- Block Search indexing with noindex — Google Search Centralaccessed on
- In-depth guide to how Google Search works — Google Search Centralaccessed on
- Learn about sitemaps — Google Search Centralaccessed on
Block revalidated on
Frequently asked questions
- What does “discovered, currently not indexed” mean?
- That the address is known to the engine but has not been downloaded yet. The problem therefore sits upstream of the page itself: its worth has not been examined. Changing its content at this stage changes nothing, since nobody has read it.
- And “crawled, currently not indexed”?
- That the page was downloaded, examined, then discarded. It is the only state that points back at the page itself. No technical setting corrects it: the question becomes what this page brings that others, already indexed, do not.
- Should you request indexing manually at every publication?
- On a normally sized site, no. Discovery is automatic and a one-off request does not buy a favourable decision. The tool is genuinely useful after a fix, to check that a previously blocked page is now readable, not to force an outcome.
- Does a sitemap solve an indexing problem?
- It helps discovery, which handles the first of the three cases. The official documentation states that it commits to neither crawling nor indexing. On a page crawled and then discarded, declaring it a second time in a sitemap changes strictly nothing.