How to Find Old and Deleted Web Pages

TL;DR

Web pages change and disappear all the time, but the old versions are often still out there in web archives and caches. If a page was edited, taken down, or a whole site vanished, you can frequently recover what it said, which is handy for research, verifying claims, security recon, and rescuing your own lost content.

• The Wayback Machine (archive.org) is the big one: billions of saved snapshots of pages over time, browsable by date.
• archive.today takes on-demand snapshots and is good at capturing pages the Wayback Machine struggles with.
• Search engine caches are essentially gone: Google retired its public cache in 2024 and Bing followed later the same year, leaving Yandex as the last major one.
• You can also save a page yourself right now so a copy exists before it changes, and there are national library web archives for deeper research.

Now for the more technical explanation

The Wayback Machine (archive.org)

The Internet Archive’s Wayback Machine is the first stop. It has been saving snapshots of the web for decades.

• Paste a URL and you get a calendar of every date it was captured; click a date to see the page as it looked then.
• Use it to see how a page (or a whole site) changed over time, or to read something that has since been deleted.
• Save Page Now: you can archive any public URL on demand, creating a timestamped copy you can cite later. Do this before content you care about might change.
• Bulk lookups: the Wayback CDX API lets you list every capture of a URL or path programmatically, which is useful for finding old pages under a site or spotting when something appeared or vanished.

archive.today (archive.ph / archive.is)

archive.today takes a snapshot on demand and stores both a rendered image and the page text. It is a great complement to the Wayback Machine because:

• It often captures dynamic or script-heavy pages that the Wayback Machine renders poorly.
• Each snapshot is a frozen, shareable, timestamped record that will not change even if the original does.
• It is handy for capturing something you expect to be edited or removed.

Search engine caches

Historically you could view a search engine’s cached copy of a page:

• Google retired its public “cached” link in 2024, so that route is largely gone.
• Bing followed in December 2024, dropping both the cached link and the cache: operator. Yandex is now the last major engine serving public cached pages, so it is the only one worth a try for very recent changes.
• Caches only ever held the latest version an engine crawled, so they are for “what did this say a few days ago,” not deep history. For that, use the archives.

Other archives and sources

• Common Crawl: a massive open crawl of the web, useful for large-scale or programmatic research rather than casual lookups.
• National and library web archives: many countries preserve their web. Examples include the UK Web Archive, the Library of Congress web archives, and national library collections (New Zealand and others). These are excellent for regionally significant or older material.
• Site-specific archives and mirrors: wikis, forums, and code repositories often keep their own history (page histories, mirrors, and forks) that predates or supplements the general archives.

Techniques that help

• Search operators: a site: search narrows to one domain, and adding distinctive phrases can surface pages that still exist somewhere even if the original link is dead.
• Guess the archive URL: Wayback URLs follow a predictable pattern, so you can jump straight to a date or ask for the closest capture to a moment in time.
• Find deleted content under a site: the Wayback CDX API can list captured paths, revealing pages that are no longer linked or have been removed.
• Recover your own lost site: if you lost a website and have no backup, the Wayback Machine’s snapshots are often enough to reconstruct the pages and content.

Limitations to expect

• Gaps: not everything is captured, and popular pages are archived far more often than obscure ones.
• Dynamic content: pages that build themselves with scripts, or that sit behind logins and paywalls, may archive incompletely or not at all.
• Exclusions: some sites ask not to be archived, or content is removed on request, so a page you want may simply not be there.
• Not evidence-grade by default: a snapshot is strong supporting material, but for formal or legal use you may need proper preservation and chain-of-custody, not just a link.

Where this is genuinely useful

• Security research and OSINT (open-source intelligence): see a target’s old pages, exposed information that was later removed, or how a site’s footprint changed over time (this pairs with the security-research tools elsewhere on this site).
• Fact-checking and journalism: show what an organisation said before it quietly edited or deleted a page.
• Recovering content: rebuild your own lost pages, or retrieve documentation and articles that have gone offline.
• Due diligence: review a company or product’s history before you rely on it.

A word on ethics and law: archives are for reading and citing what was already public. Use them for legitimate research, journalism, recovery, and authorised security work, and respect copyright and privacy in what you do with the results.

The short version

Deleted rarely means gone. Start with the Wayback Machine for history, reach for archive.today for on-demand and dynamic pages, try Yandex’s cache, the last one standing, for very recent changes, and fall back to Common Crawl or national web archives for depth. Save pages yourself when you want a copy that will still exist later, and remember the gaps and the ethics.

Related

• OSINT and recon web tools worth bookmarking

Written by Tom Langston, IT Infrastructure and Cybersecurity.