Archived and Cached Copies

The part nobody warns you about beforehand

Either you got the page taken down and then found it again — same words, same headline, sitting in an archive with a date stamped on it — or you have not asked for the removal yet, and somebody should tell you this before you pay for one. A removal does not reach the copies.

This is the practical sting in the whole subject, and it is almost always explained afterward rather than beforehand. Deleting a page at its own address does nothing to a copy sitting somewhere else. The copy is a separate page, at a separate address, held by a separate organization with its own rules or with none at all.

The honest position here sits in the middle. You can ask, and asking sometimes works. The archive that matters most publishes a request route, it is free to use, and it states in writing that it makes no promises about the outcome. Nobody can tell you in advance which way a request will go — and that is still more than exists on some of the other surfaces here, where there is nothing to ask.

What a cached copy was, and where it went

For years every page in a search engine's index came with a second, quieter version: a snapshot of the page as the crawler last saw it, behind a small Cached link beside the result. A page deleted at source could still be read through that snapshot, so getting rid of the cached copy was a separate step in any removal job — one that people who were sold removal often discovered late.

That link was retired from search results in early 2024. I have to flag the sourcing, because you deserve to know which facts here come from an operator's own documentation and which do not. The change was announced by the search engine's public liaison in posts on social media and reported widely at the time; the liaison said the company had decided to retire it and that the cache: operator would follow. No help page or blog post from the operator announcing it could be found. Treat the date and the wording as press-reported.

What remains at the search engine is narrower than people think: a free tool for asking it to refresh a stale result after the source page has already changed. It removes nothing. Filed against a live, unchanged page it does nothing at all, which is worth knowing if a service has offered it to you as a deliverable.

The archive that actually matters now

With the cached copy gone, the public web archive is the copy that counts. It is large, it has run for decades, it is free to read, and its addresses are permanent and shareable. Its captures carry dates, which is why they get used as evidence — by you, and against you.

Two ways a page ends up there. It gets crawled, without anybody asking. Or somebody saves it deliberately: the Wayback Machine help pages, read August 15, 2026, document a Save Page Now feature letting any member of the public archive a page on demand. So a capture may exist from before a page was edited, and from the day before it was deleted.

It also may not exist. Not everything gets captured, and a page can be unavailable for several reasons. So before you ask anyone to take anything down, look. Put the address of the page that is hurting you into the archive and see what is there and on what dates. That list is what any removal has to cover, and building it costs ten minutes. A removal negotiated without it solves the visible half of the problem.

What the archive publishes about taking something out

There is a real, documented route, and it is a plain email rather than a form.

“If you would like to submit a request for archives of your site or account to be excluded from web.archive.org, send us a request to [email protected]”

“This will initiate a review by our team. We do not make any guarantees beforehand about the outcome of a request.”
— Internet Archive Help Center, “How do I request to remove something from archive.org?” read August 15, 2026

The archive asks a request to include the address or addresses of the material, the time period you want excluded, the time period during which you had control of the site, and anything else you think would help.

Read that second quotation as the archive being straight with you rather than as bad news. It is one of the few operators in this subject that tells you, before you spend a minute, that the answer may be no. It publishes no decision time either. So: a route, a review, a person at the end of it, and no promise. That is the accurate description of this surface, and the reason nobody can sell you certainty on it.

The problem with that request when the page is not yours

Go back to the third thing the archive asks for: the period during which you had control of the site. The request is framed around the person who ran the website. That is a sensible way to build a process, and it does not describe you.

If you are simply somebody written about on another person's site, you do not fit that framing, and the archive publishes no separate route for the subject of somebody else's page. That is not a reason to skip writing. It is a reason to write knowing which door you are knocking on, to say plainly who you are and what the material is, and to expect a decision you cannot appeal.

Two categories do get a route aimed at the person in the material. Non-consensual intimate imagery goes to a dedicated address, and the archive lists what such a complaint must contain: identification of the material and its addresses, contact details, a statement of good-faith belief that the depiction was not consensual, and a signature. Copyright claims run through its takedown process, which turns on who owns the material — usually the photographer, not the person photographed.

Whether either applies to you is a legal question and an attorney's job rather than mine. What I can tell you is that the routes exist and what they are for. It is the pattern everywhere in this subject: a statute behind the category buys a real process, and everything else gets a mailbox.

The robots.txt advice you are going to be given

Sooner or later somebody will tell you the fix is a one-line file: get the site to add a robots rule and the archive drops the copy. The archive contradicted the general form of that publicly, and the post is still up.

“We see the future of web archiving relying less on robots.txt file declarations geared toward search engines, and more on representing the web as it really was, and is, from a user's perspective.”

“Please know that site owners can always write to [email protected] and request that content from a site be removed from the Wayback Machine and from future crawling. We process requests like that every day.”
— Mark Graham, Director of the Wayback Machine, Internet Archive Blogs, April 17, 2017, read August 15, 2026

At the same time, its current help pages still list a robots file on the site, or a site owner's direct request, among the reasons a page may be unavailable. Both things are true at once, which is why this gets argued about.

The honest statement: a robots rule may or may not suppress an archived page, and the route the archive itself describes as running every day is the email. Anybody charging you for a line in a text file is selling the weaker half of that, and selling something only the site owner can do.

The tool that works for you and against you

Save Page Now — the button that lets anyone preserve a public page on demand — cuts in both directions, and both matter to you.

For you. It is the cheapest evidence preservation there is. A dated capture in a public archive is worth more than a screenshot on your own laptop, because it is a third party's record with a third party's timestamp rather than an image you could have made in a text editor. Save it before you ask anybody for anything. Pages get quietly edited during a negotiation, and the version you complained about may not be the version anybody can see afterward.

Against you. The same button lets a hostile party preserve a page minutes before it comes down. If you are negotiating a deletion with somebody who does not want to delete it, assume the possibility that a permanent copy is being made while you talk.

None of that argues against pursuing a removal. It argues for finding the copies first, deciding what a removal is worth once you know what survives it, and hearing all of it before money changes hands rather than after.

The other copies, and the one with no door at all

The public web archive is not the only place a deleted page lives on.

archive.today, also seen as archive.ph and archive.is, is frequently the copy that outlasts everything else. I could locate no published removal policy, no contact route, and no ownership information for it. So I am not going to tell you how to get something taken off it, because there is nothing published to point you to. If somebody tells you they have a contact there, ask them to show you the page that says so.

Beyond those two: institutional and library archives, scraped mirrors, screenshot aggregators, reposts on social platforms. There is no list and no shared policy. Each is a separate publisher with its own rules or none, and the effort a copy takes has nothing to do with how much it is hurting you.

One thing I will not tell you is how often archived copies turn up in a search for a person's name. No data either way, and both stories get told confidently. What can be said is narrower: the realistic harm is that somebody already looking, who knows where to look, finds it — a different problem from a bad result sitting on the first page of your name.

What this changes about paying for a removal

Three things belong in the conversation before money changes hands, not after.

• Removed from a search engine is not gone. The page is still at its own address, a capture is still at the archive's address, and everyone holding a link still arrives.
• A background check is not a name search. Screening companies buy from data suppliers and pull records directly, so suppressing a search result does nothing to a records-based report. Those reports are governed by federal law with its own dispute process. Where that is your real problem, it is a matter for a lawyer or for that process, and not for me.
• Archives make a deletion partly reversible for the other side. Anybody selling you a removal should be telling you that at the start.

What the work looks like: find the copies before anything is negotiated, preserve what is live with dated captures, write to the archives that publish a route and say plainly who you are, and be straight about the copies that have no route at all. Where a removal is worth doing, it is worth doing with the copy list in hand. Where the copies are the whole problem and nobody publishes a way to reach them, hearing that today costs a great deal less than finding it out after paying to have the original taken down.

Frequently Asked Questions

Published 2026-08-15