How to remove or deindex content from Google: a step-by-step guide

You deleted a page, or a client asked you to, and it still shows up in Google weeks later. It is almost never because Google is slow: usually the wrong tool was used to remove content from Google.

There are three paths to deindex a page, and choosing badly costs you weeks. Here is which one applies in each case and the limits Google documents.

What "remove content from Google" means (and what it does not)

Deindexing means removing a URL from the results; deleting means removing the file from your site. Google can only do the first one: its documentation on removals notes that "Google can only remove things from Google Search results." The page disappears from search but stays on the server and in other search engines.

Three paths: temporary removal (only if you own the content), outdated content (third-party pages that are gone or changed) and personal content policies (the data is yours, the site is not). And a warning: if the content brings traffic, the question is not how to delete it but whether you should (content pruning and traffic drop).

A 30-second decision map

1. Is the property verified in Search Console? Without it, the Removals tool does not exist for you.
2. Is the content still live? Yes: temporary removal. Gone or changed: outdated content.
3. Is it your personal information, wherever it sits? Then the path is different and does not depend on the property.

The Removals tool: fast and temporary

This is the way to get something out of results today. According to its Search Console help page, it blocks "pages from Google Search results on sites that you own." It has two request types: Temporarily remove URL, which blocks the URL "for about six months" and clears the description snippet, and Clear snippet in search, which only clears the snippet until the next indexing.

Scope is either the exact URL —parameters included; anchors are ignored— or the prefix, which blocks everything starting with it. A detail that surprises people: blocking https://www.example.com also blocks https://example.com (all four www and protocol variations). It does not cover subdomains.

Two limits: a removal does not prevent crawling, only the appearance in results; and if the URL returns 404 or 502/503, Google assumes the page is gone and the request expires.

Request statuses

Five statuses: Processing request, Temporarily removed, Removal expired, Cleared and Denied.

The tool alone does not work

The help page is blunt: "The Removals tool is only one step in this process to remove a URL permanently. Using the tool alone won't work."

Permanent deindexing: the three approaches that work

1. Remove or update the content. It is the safest route and the only one that works beyond Google: "the most secure way to prevent your information from appearing in other search engines."
2. Password-protect the page, when it must stay available to specific people and out of search engines.
3. Add noindex, when it must stay public but unindexed. The technical explanation is in the guide to meta robots and noindex.

robots.txt does not deindex: the most repeated mistake

The documentation states it in bold: "Don't use robots.txt as a way to block your page." The reason is mechanical: robots.txt controls crawling, not indexing. If Google cannot crawl the URL, it cannot read a noindex either or see a 404, so the page can stay in the index indefinitely. If you blocked the page before removing the content, the rescue is to unblock, wait and block again (guide to robots.txt).

Deleting the page and returning the right status code

When you delete the file, the server should answer 404 or 410, which tells Google the disappearance is permanent; PDFs must actually be removed from the server. If the page is already gone there is nothing to request: "There's no need to request an urgent update" (nuances in the guide to 404 and soft 404 errors).

Mistakes that ruin a deindexing

Three slips repeat: using a removal to pick the good version of a duplicate, when it can remove every variation and what you needed was a canonical; treating a temporary removal as final, since it returns after six months; and forgetting URL variations (example.com/puppies, /PUPPIES and ?pet=puppies are three URLs).

Content you do not control: refreshing the result

For third-party pages and images there is the Refresh Outdated Content tool, which does not require verifying ownership and applies when the content "no longer exists or is significantly different from the current version." It does not work if you think the information is wrong or illegal. If the page is gone, "it will no longer appear in Search results"; if it changed, the snippet is removed until the crawler's next visit. If the content stays on a third-party site: contact the owner, file a copyright or trademark complaint, or use the path against sites that charge for removal.

Images and documents: their own rules

For an image that has to go today, the Removals tool works with the same clock: "the images may resurface... once the removal request expires." The permanent route, in the guide to removing your own site's images, accepts two formats: robots.txt with User-agent: Googlebot-Image and Disallow: /images/dogs.jpg, or an X-Robots-Tag: noindex header. Googlebot has to crawl the URL to read the headers, so implementing both methods at the same time makes no sense.

Redacting a document is not hiding the text

Covering a detail with a black rectangle or painting it the background colour does not redact it: as the documentation on redacted information explains, those methods "don't actually redact material in a way that prevents search engines from indexing it", because screen readers and text recognition read what sits underneath. What works: edit before publishing, crop the original image and strip the metadata.

Personal information: the path that does not need you to control the site

When the content is yours but the site is not, the personal content policies allow you to request the removal of your address, phone or email, official identification, bank or card numbers, photos of your signature or ID, medical records and credentials. The process requires the exact URLs: "Only those URLs are reviewed for removal from Search results."

The result can be full or partial: with a partial removal the site "won't show up on Google search results for any search query that contains your name or identifier", but it may still appear for other searches. Public interest weighs in, and for results from news or government sites the removal button may not even exist. Recent change: since February 2026 the "Results about you" tool can request the removal of results exposing official identification such as a driver's licence or passport; the official announcement describes a rollout that starts in the United States, so it is not accurate to promise availability elsewhere. To see what is indexed, the site: operator (search operators).

What none of these tools fix

1. They do not clean crawl errors or old 404 URLs: those drop out on their own once crawled.
2. They do not pick the "good version" of a duplicate: they can remove every variation of the URL.
3. They do not prevent crawling of the page, only its appearance in results.
4. They do not delete from other search engines or the internet: that is settled on the source site.
5. They are not a "start from scratch" reset nor a substitute for a reconsideration request.
6. Turning a site off is not removing it: recovering from a complete removal has no fixed timeline; for a temporary closure, limit functionality and return 503 if something is urgent.
7. They do not replace a business decision: if the page has links and traffic, that is decided with data.

Checklist: removing content from Google without wasting time

1. Define what goes: one URL, a folder, or everything containing a detail.
2. Check whether the property is verified in Search Console.
3. Confirm whether the content is still live: yes, temporary removal; no, outdated content.
4. If it is your personal information, go through the personal content policies.
5. File the temporary removal today and make it permanent tomorrow: remove or update, password or noindex.
6. Do not use robots.txt to block indexing.
7. Cover every variation: capitalisation, protocol, www and parameters.
8. Do not touch what crawls well: existing 404s resolve themselves.
9. Write down what you requested and when, and verify with URL Inspection.

Google documents three paths and each one has its own tool: the six-month temporary removal is a patch, permanent deindexing requires touching the site, and personal information requests are the only route when the site is not yours.