Skip to the article
Unpublish

Unpublishing, anonymisation, correction and de-indexing

Unpublish content: who can take a page down and what they can agree to

To unpublish content, you ask whoever controls the page: the publisher first, then the platform that hosts it, a search engine or a court. Each can agree to something different, from full removal to anonymisation, a correction or de-indexing. Unpublish prepares that request, files it through the proper channel and tells you plainly who decides.

Local headline abouta named personA. N. Otheranonymiseadd correctionnoindex
An illustration, not a real article. The four marks are the four changes a publisher can agree to: unpublish, anonymise, correct and noindex.

Who can actually unpublish a page?

Four parties can make a published page disappear or change, and their powers are not the same. Most cases begin with the first.

What can a publisher agree to, and what does each outcome change?

Four outcomes are on the table in most newsrooms, and a fifth exists that only a search engine can grant. A request that names the right one is far easier to say yes to.

Outcomes of an unpublish request, by what readers and search engines see afterwards
Outcome What changes for readers What changes in search
Unpublish The page is withdrawn and its address returns an error or a redirect. Copies on other sites are untouched. The result drops out once the engine recrawls the address.
Anonymise The article stays, with the name replaced by initials or a description such as "a 34-year-old man". The page still ranks for its subject but should stop appearing for your name after a recrawl.
Correct or update The text is amended, or a dated note is added: charges dropped, conviction quashed, figure corrected. The same result remains, now carrying the corrected account.
Noindex Nothing. The page stays readable in the publisher's archive and through any direct link. Engines that honour the tag drop the page from results for every query.
Search delisting Nothing. This one is granted by the search engine, not the publisher. Under the European right to be forgotten, the result is hidden for searches of your name only.

Which door fits an article, a court report, a mugshot page or a forum thread?

News articles and blog posts. The request goes to the editor or the author. If the story is wrong, the Editors' Code enforced by IPSO in the UK requires a significant inaccuracy to be corrected promptly and with due prominence. If the story is accurate but old, the argument is about proportion: a minor matter, long concluded, that still leads the results for a private person's name. That is the case to unpublish an article or, more often, to anonymise it.

Court reports. Fair and accurate reports of open court are the material editors defend hardest, and a request to delete one usually fails. A request to add the outcome (acquitted, charge withdrawn, conviction overturned) or to anonymise after a long interval is one a reasonable editor will consider, provided you can supply the court document that proves it.

Mugshot and arrest-record pages. These sites copy booking records, and some charge to take a page down. Do not pay. Google will consider removing such a page from its results when you are the subject of it, the site does not host business reviews and it demands payment to take the page down. That clears the search result, not the page.

Forum threads and comments. The author can delete their own post, moderators can remove it under community rules and the platform can remove it under its site-wide rules. Reddit's rules, for instance, prohibit revealing someone's personal or confidential information. A thread that is merely critical of you breaks no such rule, and we say so before you spend anything.

What do editors usually refuse?

An honest review starts with the requests that newsrooms turn down as a matter of routine.

How does an unpublish request go through us?

Every case passes through the same five stages. The decision at stage four is never ours.

  1. The link reaches us

    Send the address of the page, say what is wrong with it and what you would like to happen.

  2. We read it as an editor would

    We check who publishes the page, what their corrections policy says, where it is hosted and whether a platform rule or a legal ground applies. Then we say which outcome is realistic, including when the answer is none.

  3. You get a price and a timeframe

    The quote covers one case: the pages, the routes we will use and how long we expect our work to run. You decide whether to go ahead.

  4. We file, the publisher decides

    We write the request in that publisher's terms, attach the evidence and send it to whoever can act on it: an editor, a platform's reporting form or a search engine's request form. Nobody outside can make that decision for them.

  5. We follow through

    If the answer is yes, we check that the change is live and ask the search engines to refresh what they show. If it is no, we report what the refusal said and whether another door is worth trying.

Where does the right to erasure apply, and where does it stop?

Article 17 of the GDPR gives people in the EU, and in the UK under its own version of the regulation, a right to have personal data erased in defined circumstances. A controller must answer within one month, a period that can be extended by two further months for complex requests.

The limit sits in the same article. Erasure does not apply where the processing is necessary for exercising the right of freedom of expression and information, and Article 85 tells member states to reconcile data protection with journalism. A newsroom can therefore lawfully decline to delete a report that still serves a public interest. An erasure request to a publisher is strongest when it asks for less: the name, not the story.

Courts have backed that middle path. In Hurbain v. Belgium, decided by the Grand Chamber of the European Court of Human Rights on 4 July 2023, a newspaper had been ordered to anonymise a 1994 article in its online archive about a fatal road accident. The Court held, by twelve votes to five, that the order did not violate the publisher's freedom of expression.

Search engines are treated differently, because they are not publishers. Google says it delists a result only for queries related to your name, on the versions of its results served in countries where European data protection law applies. It weighs the public interest against your role in public life, and it takes into account that a news site's journalists judged the content worth publishing.

What does a noindex request achieve that deletion does not?

A noindex tag is one line in a page's code asking search engines to leave that page out of their listings. Google's documentation says that when its crawler next sees the tag it will drop the page entirely from its search results, even if other sites link to it.

Editors who will not unpublish often accept this, because the record stays intact while the page stops following a private person around. Two cautions belong in any noindex request. The page has to stay crawlable, since a crawler blocked by robots.txt never sees the tag. And the effect is not instant: Google notes that it may take months for its crawler to revisit a page, although the site owner can ask for a recrawl.

What will Unpublish not do?

Some requests we turn down, whatever the fee.

What decides the price and the timeframe of a case?

We quote each case after reading it, because two requests that look alike can differ completely in effort. The quote turns on things you can see for yourself:

The timeframe we give covers our own work. How quickly an editor, a platform or a search engine replies is theirs to decide, and we do not promise an outcome on their behalf.

What do people ask us before sending a request?

Can you remove a news article from the internet completely?

Only the publisher can remove the original, and no one can promise the removal of every copy. Syndicated versions, archive snapshots and reposts are separate pages with separate owners, and each needs its own request.

Is it better to ask for an article to be unpublished or anonymised?

Ask for the smallest change that solves your problem. If the harm is that the article appears when someone searches your name, anonymisation or noindex solves it and leaves the record intact, which is why editors agree to those more readily.

How do I get an article unpublished if the charges were dropped?

Send the publisher the document that shows the outcome and ask for one of two things: an update added to the report or the removal of your name. A demand to erase the original report, which was accurate on the day it ran, is usually refused.

Does the right to be forgotten work on newspapers?

Partly. A publisher in the EU or the UK must answer an erasure request, but journalism is protected and the publisher may lawfully refuse. The route used more often is delisting by the search engine, which hides the result for searches of your name and leaves the article online.

Should I pay a mugshot website to take my photo down?

No. Paying rewards the practice and does nothing about copies on similar sites. Google has a removal policy for content on sites that require payment for removal.

The publisher removed the page. Why does it still show in Google?

Search results are a copy taken at the last crawl, so the old title and snippet can linger until the engine revisits the address. Google built its Refresh Outdated Content tool for this situation: the page has gone or changed and the listing has not caught up.

Which official sources does this page rely on?

  1. EUR-Lex: Regulation (EU) 2016/679 (GDPR), Articles 12, 17 and 85
  2. European Court of Human Rights: Hurbain v. Belgium, Grand Chamber judgment of 4 July 2023
  3. Google Legal Help, overview of the right to be forgotten
  4. Google Search Central documentation on blocking indexing with noindex
  5. Google Search Help page on exploitative sites that charge for removal
  6. IPSO: Editors' Code of Practice, Clause 1 (Accuracy)

Unpublish works independently. It belongs to no newspaper, platform, search engine, regulator or court named here, and it holds no power over their decisions. Nothing on this page is legal advice.

Send us the link