How do I remove pages from the search?

Removing a page from search is a two-step operation, and skipping the first makes the second fail. Prohibit indexing first, then ask Yandex to hurry.

Step 1 — prohibit

If the page is gone from the site: Disallow in robots.txt, or a server response of 404, 403 or 410. A 301 is recommended alongside, for the sake of users following old links.

If the page exists but should not appear: the robots meta tag with noindex, or one of those status codes. Do not block such a page in robots.txt — the robot cannot then read the noindex.

Each method removes the page from the database within a week of the robot noticing. One caveat applies to robots.txt alone: a blocked page may still appear in results occasionally, for instance when other sites link to it. For genuine confidentiality, blocking is not sufficient.

Step 2 — accelerate

Tools → Remove pages from search results, either by URL (up to 500 per site per day) or by prefix (up to 20 prefixes per day) for a directory, the whole site, or URLs with parameters. The tool works without verification for single URLs; prefix removal requires a verified site.

The condition: the robot removes pages submitted here only if Disallow is set for them in robots.txt. This is why the order matters.

The statuses

• In the delete queue — the robot is checking the response and the prohibition.
• In progress — checked; removal within 24 hours.
• Deleted.
• Rejected — the page is still allowed for indexing, or the server answers something other than 404, 403 or 410.

Putting a page back

Remove the prohibition. The page returns when the robot next crawls and notices — which may take up to three weeks. Removal is fast and reinstatement is slow, which is the asymmetry to weigh before using the prefix option.