How to Use the Wayback Machine to Fix Dead Links
When a page you linked to dies, the content usually isn't lost — it's archived. The Wayback Machine, run by the nonprofit Internet Archive at web.archive.org, has been photographing the web since 1996 and holds hundreds of billions of page snapshots. For anyone maintaining a site, that turns a dead link from a deletion problem into a lookup problem: find the snapshot, link it, move on. Here's the whole workflow, including the parts people get wrong.
What the Wayback Machine actually is
The Internet Archive continuously crawls the public web and stores copies of pages as snapshots — each one stamped with the exact date and time it was captured. Popular pages may have thousands of snapshots across decades; obscure ones might have two. Anyone can browse them free, no account needed, and anyone can add a snapshot of a live page on demand. It is the closest thing the web has to an undo button — the standard defense against link rot, which erases roughly 38% of pages per decade.
Finding a snapshot of a dead page
- Get the exact dead URL. Copy it from your page — the full address, not just the domain. (If you're cleaning a whole page, a one-click scan gives you every dead URL in a list you can work through.)
- Paste it into web.archive.org. The search box accepts full URLs. You'll land on a calendar view: years across the top, capture dates circled below.
- Pick the right date, not the latest. This is the step people get wrong. Choose a snapshot from around the time you originally linked the page — the latest capture may show the site's death throes: a parking page, a "we've moved" stub, or a hacked version. The circled dates are color-coded; blue captures are normal 200 responses, which is what you want.
- Verify the snapshot renders. Open it and check the actual content is there — not a cookie wall or a half-loaded frame.
- Copy the snapshot URL. It has a stable format you can trust in a citation:
https://web.archive.org/web/20230415000000/https://example.com/page
— the timestamp pins that exact capture. Dropping the timestamp (or using/web/*/…) links the calendar instead of the content; don't ship that to readers.
https://web.archive.org/web/ followed by the full dead URL — then adjust the date from there.Link the archive — or use it as research?
Finding the snapshot is half the decision; the other half is what to do with it (the full decision tree is in how to fix broken links):
- Link the snapshot directly when the reference matters as it was: citations, quotes, statistics, legal or historical material. The archive preserves exactly what you cited — immune to both death and content drift, where a live page silently changes under your link.
- Use the snapshot as research, link something live when readers need a current resource: read the archived copy to understand what died, then find today's equivalent — the content's new home, or a better modern source. This is also the core research move in broken link building, where the archived page tells you what replacement to offer.
- Do neither if the reference no longer earns its place — unlink the text.
Archive pages proactively (the habit that prevents all of this)
The Wayback Machine's Save Page Now feature captures any live page on demand: go to web.archive.org/save, paste the URL, done — the snapshot exists from that moment. For anything you cite that must stay verifiable — sources in research, terms you're relying on, a competitor claim, a small personal site that could vanish — save a snapshot on the day you link it. Future-you inherits a guaranteed archive from exactly the right date instead of hoping a crawler passed by.
What the Wayback Machine can't do
- Not everything was captured. Small or short-lived pages may have zero snapshots; there is no copy to recover.
- Some owners remove archives. Site owners can request exclusion, and excluded domains show no captures even if they were once archived.
- Dynamic pages archive poorly. Heavily JavaScript-driven apps, paywalled content, and logged-in views often produce broken or empty snapshots — always open and verify before linking.
- Media can be missing. A snapshot may have the text but lose images or embedded files that lived on other rotted domains.
When the archive comes up empty, fall back to the other repair options — a live equivalent elsewhere, or removing the link (the full playbook).
The workflow in one pass
| Step | Action |
|---|---|
| 1 | Scan the page — get the list of dead URLs (extension, one click) |
| 2 | For each dead URL: check web.archive.org for a snapshot near your original link date |
| 3 | Snapshot good → link it (timestamped URL) or use it to find a live replacement |
| 4 | No usable snapshot → replace with another source or unlink |
| 5 | Rescan the page to confirm it's clean; Save Page Now for fragile new citations |