The old article about you that will not go away

The email usually starts the same way: a link, a date years in the past, and the sentence "can you get this taken down." The honest answer, most of the time, is no. Not because nobody tried, but because deletion of a published news story is not a service that exists in the form people imagine. Understanding why changes what you do next, and what you do next is usually more useful than the deletion would have been anyway.

Start with the strongest version of the other side, because it deserves to be heard first: sometimes the article really is unfair, or wrong, or written about a person who was twenty-two and is now forty-one and has spent nearly two decades being someone else. That feeling is legitimate. None of what follows disputes it. It just describes what a newsroom will actually do about it, which is narrower than most people expect, and a strategy that works within that narrowness rather than against it.

What a newsroom will actually do

Outlets have a small, fairly consistent menu, and knowing which item you are asking for changes how the request is received.

A correction fixes something factually wrong at the time of publication — a misspelled name, a wrong job title, an allegation attributed to the wrong person. This is the request newsrooms are built to say yes to, because correcting errors is part of how they understand their own credibility. Bring it with a specific, checkable claim: not "this makes me look bad" but "this sentence says X, and X is not true, and here is the record that shows it."

An update note is different: it does not change what the original piece said, but it adds a line, usually at the top or bottom, noting what has happened since — charges dropped, a role left years ago, a company since closed. This is the second most winnable request, and it is the one most people should be making instead of asking for removal, because it does the actual job: it tells a reader who arrives today that today is not the same as the date on the byline.

A deindex request asks the search engine, not the publisher, to stop surfacing a page for certain queries. This exists for narrow, specific categories — content that violates the platform's own policies, certain kinds of personal information, material a court has ordered removed — and it does not touch the article itself, which usually stays published and reachable by anyone who has the direct link.

Removal — the article coming down entirely — is what people actually want and what they are least likely to get. Newsrooms resist it on principle as much as policy: an archive that can be edited on request stops functioning as a record, and every outlet with a real editorial process knows that one removal on request becomes precedent for the next hundred. Removal happens mainly when the piece was published in clear error, defamatory, or the outlet itself decides — for its own reasons, unconnected to your request — that it no longer stands behind the story.

Where a legal request actually has standing

The honest geography of this is smaller than most advice implies.

Inside the EU and UK there is a real mechanism, generally referred to as the right to be forgotten, that lets an individual ask a search engine to delist certain results for searches on their name. It is applied case by case, weighed against public interest, and it is a delisting of the search result — the underlying article is not deleted and remains reachable through the publisher's own site, other search engines, or a direct link. It is also, notably, not something that reaches results outside the jurisdiction that granted it; a delisted result on google.de can still surface elsewhere.

Outside that jurisdiction, there is no equivalent standing right. What remains is ordinary defamation law, which requires the material to be false and damaging — accurate but unflattering reporting does not qualify, however much it hurts to read. Pursuing a defamation claim against a publisher is slow, expensive and outcome-uncertain even when the claim is sound, and the calculation should include not just legal cost but the publicity a lawsuit itself generates. A cease-and-desist letter sent to a small outlet with no legal budget sometimes works simply because responding costs more than compliance. Sent to an outlet with in-house counsel, it mostly does not, and it signals that you are contesting the story rather than correcting a fact, which moves the conversation from the corrections desk to the legal desk and slows everything down.

Displacement is the strategy that actually works, and it is slow

If removal is off the table for most people, the working alternative is displacement: getting enough current, accurate, well-linked material about you into search results that the old piece stops being the first or only thing a visitor sees. This is not a trick. It is the same mechanism covered in more detail in what people find when they search your name — a first screen has a fixed number of slots, and every credible new result competing for one pushes something else down.

The honest timeline is months, not weeks. A new page is typically indexed within days of existing and being linked from somewhere already indexed, but indexing is not ranking. What actually moves a result up a competitive query for a personal name is other pages linking to it and a track record of people clicking it when they search that name, and both take time to accumulate because personal names are searched far less often than the queries most sites optimize for. Expect a long flat period and then, if the material is genuinely good and genuinely linked, a jump rather than a steady climb.

This means starting before the situation is urgent. Someone who builds their displacement material the week before a job interview is doing the right thing at the worst possible time — the material will still be unranked when it is needed. This is a maintenance habit, not a crisis response, and the people who handle an old article well are usually the ones who had already been quietly adding current material for a year before anyone searched their name in anger.

Writing your own account, without relitigating the story

One piece of displacement material deserves particular care: a page, under your name, giving a factual account of what the old article covers. Done well, it becomes the result a reader compares the old piece against, in your own words, without you needing to be in the room to explain yourself.

The failure mode is writing it as a rebuttal. A page whose entire purpose is arguing with a specific old article confirms to any reader that there is something worth arguing about, and raises the salience of the very piece you are trying to move down, because now there are two pages about the same incident where there used to be one.

The version that works is shorter, plainer, and does not mention the article at all. State what happened in one or two factual paragraphs, in the same unremarkable tone you would use for any other part of your professional history — no defensiveness, no "as has been misreported." The goal is a paragraph a stranger reads once and moves past, not an essay that argues its case. If the underlying situation is closed — charges dropped, the company dissolved, the dispute settled — say that plainly and let it sit there as a fact.

The Streisand risk

There is a well-documented pattern where the attempt to suppress something draws more attention to it than the thing would have received on its own, named after a 2003 incident where a lawsuit intended to remove a photograph resulted in far more people seeing it than ever had before. The mechanism is straightforward: legal threats, takedown notices and public complaints are themselves newsworthy in a way the original material usually was not, and journalists find "subject tried to suppress this" a more compelling story than whatever the original piece said.

The practical guardrail is proportion. A polite correction request to a corrections desk carries essentially no Streisand risk — it is a routine, private interaction that nobody writes about. A public callout, a legal threat to a well-resourced outlet, or a campaign enlisting others to demand removal carries real risk, because each escalation is itself an event, and events get covered. The more public and forceful the request, the more it needs to be justified by something genuinely serious, not merely unwelcome.

When to say nothing at all

The hardest advice to take is also often the correct one: for a lot of old, low-visibility coverage, the right move is no move. An article that ranks on page three of your name's search results, that nobody currently searching for you is likely to reach, does not need a correction request, an update note, or a displacement campaign. Requesting anything draws your own attention to something quietly fading on its own, and outlets sometimes notice a sudden inquiry about an old piece and choose, on their own initiative, to refresh or re-promote it — the opposite of what you wanted.

The situations that genuinely warrant action are the ones doing active harm right now: a piece that ranks on the first screen for your bare name, that a specific person you know is about to search — a hiring manager, a client, an investor — and that misrepresents something in a way you can point to specifically. Everything else is better handled by the slow, unglamorous accumulation of current material described above, the same habit that helps with a shared name in what to do when you share a name with someone else, and increasingly by making sure the answer a language model gives about you is as accurate as the one a search engine gives, which is its own separate project covered in what AI assistants say about you.

None of this is satisfying in the way "get it taken down" would be satisfying. It works anyway, on the timeline these things actually run on, which is longer than anyone wants and shorter than doing nothing.