On this page
Reachable, but not in Google
With noindex you keep a page out of search results on purpose, while it stays reachable for anyone who has the address or follows a link. Google is strict about it: such a page drops out of the results entirely, inbound links or not.
You set the rule in one of two ways, with the same effect:
- As a meta tagin the head of the page: meta name="robots" content="noindex". The usual route for a web page.
- As an HTTP headeryour server sends: X-Robots-Tag: noindex. The only route for files without a head, such as a PDF, an image or a video.
The same tag takes other rules too. Nofollow asks crawlers not to follow the links on the page, none means noindex and nofollow together, nosnippet forbids a text snippet under your result. With conflicting rules the most restrictive wins.
The robots meta tag was already in the W3C's HTML 4.01 standard of December 1999. For years site builders also put noindex in robots.txt; Google never documented that and has ignored it since 1 September 2019.
Noindex plus robots.txt: the mistake that flips the result
Anyone who wants to hide a page twice over adds noindex to the page and also blocks it in robots.txt. The opposite happens: Google may not visit the page, so it never reads the noindex, and it can keep showing the address if other sites link to it, without a description. Pick one. To get a page out of Google, let Google visit it and set noindex.
Google has to read the page first
<meta name="robots" content="noindex">X-Robots-Tag: noindex
Crawl allowedworks
- robots.txt allows the addressGooglebot may fetch the page.
- Google reads the pageThe head or the header says noindex.
- Out of the resultsOn its next visit, Google drops the page.
Crawl blockedtrap
- robots.txt: DisallowGooglebot may not fetch the page.
- Google never reads the pageThe noindex is there, but nobody sees it.
- The address can stayIf other sites link to it, Google shows the address without a description.
address stays visible
Which pages belong out, and which do not
Noindex is for pages that exist for people already on your site, not for people searching.
| Page | Noindex? | Why |
|---|---|---|
| Thank-you page after a form | Yes | No search value, and visitors arriving from Google count as fake leads in your analytics |
| Internal search results | Yes | Endless variations of the same content |
| Login, basket, account | Yes | Personal and without search value |
| Staging or test site | Yes, ideally with a password | A password also keeps people and other crawlers out |
| PDF price list you do not want in Google | Yes, with X-Robots-Tag | A PDF has no head for a meta tag |
| Product temporarily out of stock | No | The page stays relevant and comes back |
| Two versions of the same page | No, use a canonical | Google advises against noindex for choosing the main version within one site |
| Page that is gone for good | No, a 301 redirect or a 404 | There is nothing left to hide |
Four tools, four different questions
Four tools, four different questions. Robots.txt is about visiting, noindex about showing, a canonical about which version counts and a 301 about moving.
| Tool | May Google read the page? | Does it show in results? | Use it for |
|---|---|---|---|
| noindex | Yes | No | Thank-you pages, internal search, accounts |
| robots.txt Disallow | No | Sometimes the address, without description | Sections Google need not crawl, such as filters |
| canonical | Yes | The version you point to | The same content on several addresses |
| 301 redirect | The old address forwards | The new address | A move or merged pages |
| Password | No | No | Staging sites and internal documents |
A noindex page does not belong in your sitemap either: that list holds only the pages you want in Google. Nor does a noindex page need a title tag or meta description written on search terms: nobody sees them in Google. If you do want to be in Google but without a text excerpt, also in Google's AI answers, the rule is nosnippet; how those AI answers are built is in our piece on Google AI Mode after I/O 2026.
A relaunch that copied the staging rule
Take a B2B software firm in Rotterdam relaunching its site. During the build, the staging site carries noindex on all 80 pages, as it should. At launch the new site is copied over, rule included. The old site earned 5,000 impressions a month in Google. Within a few weeks that sinks close to zero: Google revisits the pages, reads noindex and drops them. Nobody notices, because the site works fine.
Removing the rule takes five minutes. Coming back takes longer: Google has to revisit every page, and Google says that can take months. With URL Inspection in Search Console you request a new crawl page by page, which speeds things up for your key pages. That is why noindex sits at the top of every checklist for moving a website to a new host.
Free to set, expensive to forget
The rule itself is free: one line in the head or a header on your server, in WordPress a tick box per page in your SEO plugin. Search Console is free and lists in its page indexing report which addresses Google excluded because of a noindex. Our SEO checker shows it per page, including when it sits in a header your browser does not show. If your site sinks without a clear reason, the website not showing on Google check starts with exactly this rule.
If a page has to go urgently, for example because it exposes personal data under the GDPR, the removal tool in Search Console hides it for about six months. That is temporary: during that time set a noindex, a 404 or 410, or a password, or it comes back. How long a normal page takes to get into Google is in how long until my website shows on Google.
Before and after a launch: seven checks
- The staging site carries noindex or sits behind a password, and that is on the launch list.
- After launch, no page that belongs in Google carries noindex, not in a header either.
- No rule in your robots.txt blocks a page you rely on noindex for.
- Your sitemap holds no address with noindex.
- Thank-you pages, account pages and internal search results do carry noindex.
- PDFs that do not belong in Google get X-Robots-Tag: noindex from your server.
- The page indexing report in Search Console lists no key page as excluded by noindex.
What site owners ask about it
What is the difference between noindex and nofollow?
Noindex is about the page itself: do not show it in results. Nofollow is about the links on that page: do not follow them. With noindex alone, Google may still follow the links on the page to your other pages.
How do I check whether a page has noindex?
Open the source and search for noindex in the head. A header does not show there: URL Inspection in Search Console shows it, and so does our SEO checker. If a page is missing from Google and the page itself says nothing, look at the X-Robots-Tag.
Can I put noindex in my robots.txt?
No. Google never officially supported it and has ignored it completely since 1 September 2019. Robots.txt only governs which addresses Google may visit. Put noindex in the page or the header, and let Google visit that page.
How fast does a noindex page disappear from Google?
Only once Google revisits it, which can take from a few days to months. Request a new crawl with URL Inspection in Search Console. If it has to go now, use the removal tool as well: it hides the page for about six months.