How to Keep a Page Out of Google Search

To keep a page out of Google, add <meta name="robots" content="noindex"> to its <head> (or send an X-Robots-Tag: noindex header), make sure robots.txt does not block it, and let Google crawl it again. Blocking a page in robots.txt doesn't hide it: Google can still list the URL, it just can't read the page. If the page is already showing in results and needs to go today, use Search Console's Removals tool as well. Anything genuinely private should be behind a password or deleted, not just hidden. Each method below takes a few minutes.

Why this is worth checking now#

On 21 October 2026 Google refreshed its page on controlling what you share with Google Search, the overview of every way a site owner can keep content out of results. Nothing in it is brand new, but it's a good reminder of which tools Google actually supports, and the run-up to Christmas is when small sites tend to collect pages they don't want found: a staging copy of the new menu, a gift voucher thank-you page, a Black Friday page that's finished, or a PDF price list from last year.

Typical pages to hide on a small-business site:

• Thank-you and confirmation pages after a form or payment, so nobody lands on "Thanks for your order" from Google.
• Staging or preview copies of the site, which duplicate your real pages.
• Your 404 page.
• Old offers and internal pages, such as a staff rota or a client proof page.

Option 1: the noindex meta tag (any site)#

Google's guide to blocking indexing with noindex says that once Googlebot crawls a page and finds the rule, it drops the page from results entirely, even if other sites link to it. Put this inside the page's <head>:

<meta name="robots" content="noindex" />

That works with every search engine that supports the rule. Use name="googlebot" instead if you only want to affect Google. You can combine rules, for example content="noindex, nofollow".

Several Freeframe templates already do this on their error page: the Atlas, Forge and Smile 404.html files carry the same tag, so you can copy the line from there.

How to add it in each stack:

• HTML: paste the tag into the <head> of each page you want hidden. Leave it out of every other page.
• Next.js: export metadata from that route's page.tsx with robots: { index: false }. The generateMetadata reference lists the robots options, and Next.js writes the matching <meta name="robots"> tag for you.
• Astro: add the tag to that page's <head>, or pass a noindex prop to your layout and output the tag only when it's set.
• WordPress: an SEO plugin usually adds a "noindex" option to each page's settings. To hide a whole site, such as a staging copy, tick Settings > Reading > Search engine visibility. The WordPress Reading Settings guide says that since version 5.3 this adds a noindex,nofollow robots tag to every page. Untick it before launch.

Option 2: the X-Robots-Tag header (PDFs, images and whole folders)#

A meta tag only works in HTML. For a PDF menu, an old price list or a folder of client proofs, send the same rule as an HTTP header. Google treats X-Robots-Tag: noindex the same as the meta tag, and its robots meta tag specification lists every rule both accept.

On Netlify and Cloudflare Pages, add a plain text file called _headers (no extension) to the folder you deploy. For example, to hide every PDF in a downloads folder and a thank-you page:

/downloads/*
  X-Robots-Tag: noindex

/thank-you.html
  X-Robots-Tag: noindex

The syntax is the same on both: a path on one line, indented headers below it. See Netlify's custom headers docs and Cloudflare Pages' headers docs. Cloudflare notes a limit of 100 header rules at the time of writing, and both say these headers don't apply to pages generated by functions or server-side rendering; those need the header set in code.

On Vercel, add the header in vercel.json or, for a Next.js site, with the headers() function in next.config.js. If your host doesn't let you set headers, use the meta tag for pages and remove or password-protect files you don't want found.

To check a header is live, run curl -I with the page's address and look for x-robots-tag: noindex in the output.