Why Google is not indexing your page

Before diagnosing anything it is worth separating two complaints that arrive in the same sentence. 'My page is not on Google' can mean the page is not in the index at all, or that it is in the index and sitting far enough down results that nobody encounters it. These have almost nothing in common. The first is usually a binary technical fault with a specific cause you can find. The second is a competitive question with no single cause.

The quickest way to tell them apart costs nothing: search for the page's exact URL, or use a site: query restricted to your domain. If the page comes back, it is indexed and your problem is position, which is a different article. If nothing comes back, it is genuinely not in the index and everything below applies. Doing this first saves a great deal of wasted effort, because the fixes for the two problems have no overlap and people routinely spend weeks on the second while suffering from the first.

Indexing is not immediate and was never promised to be. A newly published page on a site that is not crawled frequently can wait days, and on a small site with little link activity it can be longer. There is no queue position to check and no service level. A page published yesterday and absent today is behaving normally, and the correct response is usually to wait rather than to start changing things.

This matters because the alternative is destructive. Someone publishes a page, finds it missing, and begins editing the title, rewriting the content, resubmitting repeatedly and adding directives — which produces a page that has changed three times before it was ever assessed once, and makes the eventual diagnosis harder because nobody now knows which state was the problem. Give a new page a few days before treating its absence as a fault. If you want to help legitimately, make sure it is linked from somewhere already crawled and listed in your sitemap, then leave it alone.

The next tier is the one people find hardest to hear: the page was crawled, assessed, and not considered worth indexing. Search engines do not index everything they find. A page with two paragraphs of generic text, a page that exists to hold a keyword, a location page identical to eleven others with the city name swapped, a product variant differing from its sibling by a colour — all of these can be crawled and passed over, and the report will say the page was discovered or crawled but not indexed.

This is not a penalty and there is nothing to appeal. It is a judgement that the page adds nothing to what is already indexed, including what is already indexed on your own site. The fix is genuinely to make the page worth a slot, which means adding what a customer would actually ask about — specifics, prices, process, constraints — rather than adding length. Padding a thin page to a word count produces a longer thin page. If you cannot state what a visitor would learn from this page that they would not learn from your other pages, the honest answer may be that the page should be merged into one of them.

Below the judgement calls sit the mechanical causes, each one binary and each one readable from the page itself, which makes them the fastest to eliminate. First, a robots meta tag or an HTTP header saying noindex. This is the single most common self-inflicted cause, and it frequently arrives from a template default, a staging configuration promoted to production, or a plugin setting somebody changed for an unrelated reason. Search your page's own source for the word noindex.

Second, robots.txt blocking the URL, which prevents fetching. Third, a canonical link pointing somewhere other than this page, which tells the engine to index the other page instead — a common accident when a template hardcodes a canonical to the homepage, or when a page was duplicated from another and the canonical came with it. Fourth, the page not returning a successful response to a crawler even though it loads in your browser, which happens with geographic or user-agent restrictions and with authentication on a page nobody realised was gated. All four are checkable in minutes: read the head, read robots.txt, read the canonical, and fetch the URL fresh in a private window.

A page nothing links to is hard to discover and harder to justify indexing. Discovery is the first half: crawlers find pages by following links, and a sitemap entry helps but is a weaker signal than a link. The second half is that a page nothing links to has nothing arguing for it — no part of your own site says this page is relevant to anything — which feeds directly into the judgement described earlier.

This is the cause most often missed, because the page is fine and the problem is entirely outside it. The page looks correct in every respect, has no bad directive, is decent content, and sits unreferenced. It happens to campaign landing pages after the campaign, to pages dropped from a simplified menu, and to anything created outside the normal publishing flow. The fix is to link to it from somewhere relevant that is already indexed, in words that describe what it is, and then wait for the recrawl rather than resubmitting repeatedly.

Search Console's page indexing report is the only free source that distinguishes these cases, and it requires a verified property, so it is free but not automatic. Its states map closely onto the causes above. 'Discovered — currently not indexed' means the URL is known and has not been fetched, which points at crawl capacity, a slow site, or a page with nothing linking to it. 'Crawled — currently not indexed' means it was fetched and passed over, which points at the thin-or-duplicate judgement. 'Excluded by noindex tag', 'Blocked by robots.txt' and 'Alternate page with proper canonical tag' each name a mechanical cause directly. The URL inspection tool will also show you the canonical Google selected, which is frequently not the one you declared.

Two honest limits. The report samples and does not list every affected URL, so an absence from the list is not proof a page is fine. And there is no facility anywhere that tells you when a page will be indexed, or guarantees that it will be — no tool has that information, because the decision is made by a system that does not publish its queue. Anything promising a timeline to indexing is describing something it cannot know. The realistic posture is to eliminate the four mechanical causes, which are entirely within your control and readable from the page, then make sure the page is linked and worth a slot, and then accept that the remaining variable is somebody else's system operating on its own schedule.