# Remove a page from archive.today and other archives

How to remove a page from archive.today (archive.ph) and other web archives: what each one allows, what it refuses, and the order of steps to follow.

---


You cannot remove a page from archive.today by deleting the original, because the archive keeps its own copy. The operator is anonymous and, as far as we could read, publishes no removal procedure, so the realistic routes are a direct report to the archive, a request to search engines to stop listing the archive address, and a legal notice where you hold the rights or a court order.

The same logic applies to every public web archive, but the details differ a great deal. Some archives publish a removal process and a review panel. Some, such as Common Crawl, honor a robots.txt rule going forward but describe no way to pull out data already collected. Others publish nothing we could read. This guide sorts them one by one, says plainly where no official process could be found, and gives the order in which to act so you do not waste weeks on the wrong door. archive.today's own FAQ could not be opened for this guide, so what we say about it comes from the English Wikipedia article on the service, a published security guide, a public forum thread and a security vendor's account of a complaint it relayed. The Wayback Machine has its own published exclusion route and is left out of the detail here.

<div class="box"><b>The short version</b>
<ul><li>Deleting or changing the original page does not touch an archive copy. It only stops new captures of the old content and lets search engines refresh their own records.</li><li>archive.today (also archive.ph and archive.is) has no removal process we could read, and one public thread we read reports unanswered requests, while one reported case shows the operator acting on a complaint. Treat a direct report as a request, not a right.</li><li>Search engines can be asked to stop listing the archive address for certain categories of content, though the page itself stays online.</li><li>Common Crawl, Perma.cc, the UK government web archive, Arquivo.pt and Megalodon each have different, partly published routes. Use the table below to pick.</li><li>A copyright notice is for rights holders only. Filing one that is not true creates a second problem.</li></ul></div>

## Why removing the original page does not remove the archive copy

An archive stores a separate copy, so the host of the original page has no control over it. The English Wikipedia article on archive.today describes each capture as a replica plus a screenshot, with CSS converted to inline styles and JavaScript-generated content frozen.

Taking the original down still matters. It stops the content being served to visitors and to new capture requests, and it lets search engines update their own records: Google's help page on outdated content says that when a page was deleted but still appears, you can use its refresh option for that page address. That tool concerns the page on the original site, not a copy at an archive address.

So there are always two layers. The source layer is the page you or someone else controls. The copy layer is the archive. They have different owners and different rules. Work the source first because it is usually fastest, then move to the copy layer knowing what you can request and what you cannot.

## How to remove a page from archive.today, and what it refuses

archive.today is a user-driven snapshot service with an anonymous operator, no published takedown process we could read, and no recognized duty to follow robots.txt. Its domains include archive.today, archive.ph and archive.is, so an address on any of them is the same service. Wikipedia reports that, according to the site's own FAQ, it acts as a direct agent of the user who asks for a capture, which is why it says it does not obey robots.txt.

### What it does not give you

- **No automatic opt-out.** Wikipedia's article says that once a page is archived, no internet user can delete it directly, and quotes the site as saying a website cannot protect itself from having a user mirror it.
- **No named counterparty.** The operator is anonymous, and the same article reports that the registrar was subpoenaed in October 2025 in an undisclosed criminal investigation. There is no company to write to and no registered agent we could identify.
- **No assured reply.** In one public forum thread, the person who started it said they had sent multiple reports and emails to the site and got no response. Nobody in the thread reported a successful removal. One thread is not a statistic, but it matches the lack of a published process.

### What does sometimes happen

The picture is not entirely closed. The security vendor AdGuard published an account of a complaint campaign against archive.today in which it contacted the operator directly. It reports a reply within hours and removal of the flagged material, and says the operator claimed it had never received earlier notices about those addresses. That case involved illegal content, so it does not show how a reputation complaint would be treated. It does show the operator reads at least some reports.

A published security guide describes a "report bug or abuse" link on each capture that opens a message form, with escalation by email if nothing happens. We could not confirm from a primary source how the form is handled. The guide also advises choosing the copyright category, which you should do only if the copyright is genuinely yours.

<aside class="callout">
<span class="callout-label">Rule</span>
<h4>Never file a copyright claim you cannot back</h4>
<p>A copyright notice is a formal statement by the person who owns the work. Using that category as a shortcut for a privacy or reputation complaint can expose you to a counterclaim and makes your next, honest request less credible.</p>
</aside>

### Stopping new captures

The same guide says archive.today ignores robots.txt and meta tags and uses no distinct user agent, so the only blocking method it describes is blocking the service's server addresses, with the warning that it can fail when new servers appear. That affects future captures only, never existing copies.

### A capture that differs from your page

Wikipedia reports that on 20 February 2026 English Wikipedia banned links to archive.today, citing a DDoS episode and evidence that archived content had been altered. If a capture shows words that were never on your page, keep a dated screenshot of the live page, the capture address and the differing text. That is an accuracy problem and may need a lawyer.

## Which route fits which archive

The right route depends on who runs the archive and what it publishes, so match the archive first and act second. The table summarizes what we could read. "Not found" means we looked and the archive's own pages did not describe a route.

<div class="tablewrap"><table><thead><tr><th>Archive</th><th>Published removal route</th><th>Who can ask</th><th>Notes from the sources</th></tr></thead><tbody>
<tr><td><strong>archive.today, archive.ph, archive.is</strong></td><td>Not found</td><td>Anyone, informally</td><td>Anonymous operator, ignores robots.txt, replies reported as inconsistent</td></tr>
<tr><td><strong>Common Crawl</strong></td><td>robots.txt for CCBot, a public opt-out ledger, contact page</td><td>Content owners</td><td>Past crawl removal not described</td></tr>
<tr><td><strong>Wayback Machine</strong></td><td>Email route for owners, copyright policy, separate intimate-image page</td><td>Site or account owners, copyright claimants</td><td>Reviewed by a team, no outcome promised</td></tr>
<tr><td><strong>UK government web archive</strong></td><td>Takedown policy and request form</td><td>Owners of archived sites</td><td>Access removed on receipt, panel decides later</td></tr>
<tr><td><strong>Arquivo.pt</strong></td><td>Signed declaration, or report of illegal content</td><td>Authors of the content</td><td>Grounded in protected rights</td></tr>
<tr><td><strong>Perma.cc</strong></td><td>Delete within 24 hours of creation, owner contact afterwards</td><td>Link creator, site owner</td><td>Copy may remain in library archives</td></tr>
<tr><td><strong>Archive-It collections</strong></td><td>Through the institution that runs the collection</td><td>Site owners</td><td>Honors robots.txt by default</td></tr>
<tr><td><strong>Megalodon</strong></td><td>Criteria published, no form or steps found</td><td>People named, owners, authorities</td><td>Judged case by case</td></tr>
<tr><td><strong>Ghostarchive</strong></td><td>Not found (FAQ could not be read)</td><td>Unknown</td><td>Treat as unverified</td></tr>
</tbody></table></div>

The rows after archive.today are expanded further down. First, the sequence that holds for almost all of them.

## The real order of operations for a copy that will not go away

Work from the cheapest, most certain step to the least certain, and keep a record at each stage. A search engine, a host or a court will ask whether you have already dealt with the source.

<figure class="fig">
<figcaption><b>Order of operations for an archived copy</b></figcaption>
<ol class="fig-steps">
<li><b>List every copy</b>Search each archive for the address of your page and for your domain, and note every capture address with its date. Each capture is a separate address.</li>
<li><b>Fix the source</b>Ask the host or site owner to delete or correct the original. This is where the process is clearest, and it stops new captures of the old content.</li>
<li><b>Ask the archive</b>Use the route in the table. Be specific: capture address, what it shows, why it should go, and what you are entitled to ask.</li>
<li><b>Ask search engines to drop the archive address</b>File a request for each archive address that appears in results, using the category that fits your content, even if the archive has not answered.</li>
<li><b>Escalate to the legal route only where it fits</b>Copyright notices belong to rights holders. Court orders and lawyer letters are for unlawful content, and an anonymous operator makes them slow.</li>
<li><b>Recheck on a schedule</b>Search again after a few weeks. A removed capture can be re-created if someone archives the page again, so check the source page stays changed.</li>
</ol>
</figure>

Keep evidence before you change the source. If you remove the page first, you lose the proof of what it said and when, so take dated screenshots and save every capture address.

## Asking search engines to stop listing the archive address

A search-engine request removes the archive address from results, not the archive copy. It often does the most for a reader's real experience, because few people reach an archive without a search result pointing there.

Google's Results about you tool covers contact details, government identifiers, financial account numbers, images of signatures or IDs, private records, login credentials and doxxing content. The help page says only the addresses you submit are reviewed, a full removal means the page no longer appears in Google Search, and a partial removal hides it for searches containing your name or identifier. It also says Google can only remove things from its own results, that the page stays on the web, and that Google may keep newsworthy content.

The page we read does not single out archive sites, so we cannot say how reviewers treat an archive address as such. The form asks for the address of the page showing your information, and an archive address is an address like any other. Check first that the capture is indexed at all: if it is not in search results, there is nothing to request.

Outside those privacy categories, search engines run legal removal processes. Google's legal removal help page says requests are made per product with the specific addresses involved, and that the content may still exist elsewhere on the web. Other search engines have their own forms, which we did not read. If the problem is an old result that keeps ranking, better pages may help more than removal (see [suppressing negative results](/suppress-negative-results)).

## When a legal route applies, and when it does not

A legal route is worth its cost only where you hold a right the operator must answer, and an anonymous operator makes every route harder. Three situations differ.

**You own the copyright.** If the archived page reproduces text, photographs or other work you created, a copyright notice is the formal tool. Rules and consequences vary by country, so have counsel review the notice. If the work is not yours, this route is closed.

**The content is unlawful in your country.** Defamation, harassment and non-consensual intimate imagery have their own laws. Arquivo.pt's help page states that for clearly illegal content posted by others, the body behind it removes or blocks access under a named Portuguese law. For archive.today we found no equivalent channel and no identifiable operator, so a lawyer would need to decide whether and where a claim can be served. Our guide to [removing negative content](/remove-negative-content) covers the route-by-route logic.

**There is a court order.** An order may be directed to a host, registrar or search engine even where the operator is unknown. A forum commenter suggested emailing the hosting provider's abuse contact and then taking legal action. Nobody reported trying it on archive.today, so test it with a lawyer rather than rely on it.

Two risks apply to every route: a heated complaint can be quoted or archived itself, and a false claim creates a second problem. Write requests in plain, factual language and never threaten.

## Common Crawl and the datasets built on it

Common Crawl lets you block its crawler going forward and appear in a public ledger of opt-outs, but the pages we read describe no way to have already collected data taken out. It is a nonprofit that maintains an open repository of crawl data, and its FAQ says the data is stored on Amazon's S3 service for bulk download.

The published mechanism is robots.txt. Its crawler page says to add a rule that names CCBot and disallows everything, and the FAQ says the crawler periodically checks whether the file changed. A Common Crawl blog post warns that blocking it can affect the discoverability of your website.

The opt-out ledger, renamed from "registry", is described as a published list of every legal opt-out request received, meant to alert downstream users to excluded material. Common Crawl says it is committed to honoring legitimate requests. Neither the ledger post nor the FAQ says whether data in earlier crawls is deleted, so treat past data as outside what a site owner can recall, and use the contact page for a specific case. We found no direct role for this data in search results, and what AI providers do with it is a separate question for each provider.

## Other public archives and how to get a page out of them

Each remaining archive publishes something different, and some publish nothing. These are the ones we could read.

**The UK government web archive (The National Archives).** Its removal page says removal is allowed only in exceptional circumstances under a takedown policy, and the requester must name the criterion. Whole pages or documents can go, but text cannot be redacted because snapshots are not editable. Public access is removed on receipt, pending review. A panel that meets every other month decides, and a refusal can be appealed to the Chief Executive and Keeper. No deadline is given.

**Arquivo.pt.** The Portuguese web archive's help page says the author of the archived content can request blocked access, and for a legal entity a representative must ask. It advises doing so only where content infringes legally protected rights, with a signed declaration on its template listing all affected addresses. It gives no processing deadline.

**Perma.cc.** Its documentation says links can be deleted within 24 hours of creation, then move to a permanent collection. Site owners who do not want archives shown publicly can contact Perma.cc, which says it will respect that, but the copy stays in library archives and off the Perma.cc site.

**Archive-It collections.** These are run by institutions, so the contact is the institution. Columbia University Libraries' web archiving FAQ says it will honor requests to remove archived content. Archive-It's support page says its crawler honors robots.txt by default and describes no owner removal process.

**Megalodon.** The Japanese service run by Affility Co., Ltd. publishes criteria rather than a process: infringement of rights, illegal information, ties to serious crimes, or obscene material, plus requests from courts, authorities and source-site operators. Requests from the person who made the snapshot are not accepted, and requests from the person whose information was exposed are weighed case by case. We found no form, steps or timeline. Wikipedia says it does not archive sites with a noindex or noarchive tag in robots.txt, which its policy page does not mention, so test rather than assume.

**Ghostarchive and small archives.** We could not read Ghostarchive's pages, so we describe no process for it. For any archive that publishes nothing, the options are a direct message to whatever contact it shows, a search engine request, and the legal route.

## A realistic timeline for one archived page

Timelines depend on the archive, the content and who answers. This scenario uses a hypothetical founder whose old profile page was captured by two archives and found in search. The ranges come from our experience, never as promises, and the only sourced figure is the National Archives panel schedule.

<figure class="fig">
<figcaption><b>One archived page, weeks one to ten</b></figcaption>
<ol class="fig-steps">
<li><b>Days 1 to 3, evidence and source</b>Take dated screenshots, list every archive address and ask the host to delete the original. The source usually changes within days if the host is reachable.</li>
<li><b>Days 3 to 7, first requests</b>Report to each archive and file search-engine requests for each archive address. Archive replies are uneven, and some never arrive.</li>
<li><b>Weeks 2 to 4, search outcome</b>Search engine reviews usually return within a few weeks as a full removal, a partial one or a refusal with reasons.</li>
<li><b>Weeks 4 to 8, escalate where it fits</b>A rights holder can send a formal notice through counsel. For the UK government archive, allow a couple of months: the panel meets every other month and no deadline is published.</li>
<li><b>Week 10, recheck</b>Search the archives and results again, record what remains and decide whether to keep pushing or to work on better pages instead.</li>
</ol>
</figure>

Some copies will not move. The honest outcome for a stubborn archive.today capture is often a copy that still exists but that no search result points to, which is a smaller problem than a copy that ranks.

## What to do this week

The best first move is a short list, not a letter. Write down the page address, each capture address, the capture date and what the page shows that you want changed. Match each archive to its row in the table, then start with the source and the search-engine request, the two steps with published rules.

<div class="box hot"><b>If this is happening to you now</b>
<p>A copy you cannot reach is a harder file than a page you can. Reputation Partners can review the capture addresses, rank the routes by what the sources support and tell you where a request is worth sending. Start with a private <a href="/audit">audit</a>.</p></div>

## Questions people ask

### Can you remove a page from archive.ph?

Not by right. archive.ph is the same service as archive.today, and we could read no published removal process. Reports to the operator are sometimes answered and sometimes not. Ask through the report link on the capture, ask search engines to drop the address, and use a legal route only if it fits your situation.

### Does robots.txt stop archive.today?

No, according to the sources we read. Wikipedia reports the service says it acts as a direct agent of the user who requests a capture and so does not follow robots.txt. A security guide says it uses no distinct user agent, leaving only the blocking of server addresses, which it warns can fail when servers change.

### Can I remove my site from Common Crawl?

You can block its crawler going forward with a robots.txt rule for CCBot, and you can ask to be listed in its opt-out ledger. The pages we read do not say whether data from earlier crawls is deleted, and they describe no takedown process. Use its contact page for a specific case.

### What if the archive copy shows something false?

Keep proof of what your live page said and when, plus the capture address and the false text. A copy that differs from the original is an accuracy problem and may be a legal one. Raise it with the operator and, where the content is false and harmful, take advice from a lawyer in your country.

### Will deleting my website remove it from all archives?

No. Deleting the site stops new captures of the old content, but each archive keeps its own copy under its own rules. Some route requests through partner libraries or institutions, and some publish nothing. Treat every archive as a separate request.
