Link Rot: Meaning, Examples and How to Keep What You Save

Link Rot: Why Saved Links Stop Opening and What to Save Instead

You saved an NYT piece three months ago. The opening paragraph stayed with you. Something about a town in Nebraska that rebuilt itself after a flood. Tonight a friend brings up something adjacent at dinner. You go to pull it up on your phone. Paywall. You read the part you needed on a friend's tab, hit save, moved on. The save took you to a wall.

You try to remember the headline well enough to Google around it. You can't. The save was the headline, more or less, and now the headline is behind the paywall too.

You close the tab. You shrug and say, "I read something about this once."

What is link rot?

Link rot is when a link stops leading to what it pointed to. The page was deleted, moved to a new address or the whole site went offline, so the link now ends at an error such as 404 Not Found. Wikipedia's link rot entry lists other names for it: link death, link breaking and reference rot.

A quieter version is content drift. The link still opens, but the page says something different now. Researchers at Harvard Law School describe content drift as unannounced changes at the same address: retractions, additions or a full replacement, in their study of links in The New York Times.

For someone who saved a page, both end the same way. You open the save, and the thing you wanted is not there.

A saved address and saved content are different

A conventional bookmark points back to a web address. It works while that address still leads to the page and while you can still access it.

The frustrating moment comes later: you remember saving the right thing, find the link, and still cannot reach what you needed. The capture step succeeded, but the return trip did not.

That distinction matters when choosing where to keep something. If the URL alone is enough, a bookmark is simple and appropriate. If the material must remain available for later recall, consider whether your save method keeps the content rather than only its address.

All the ways a save can stop working

Walk through your own saves and you will recognize most of these.

The paywalled article. The first read was free, or a friend's tab was logged in, or a Twitter link broke the wall once. You saved it. You come back. The wall is up. No preview of the line you cared about. The save is now a recommendation to subscribe.

The deleted YouTube. Creators delete videos. Channels get banned. Music videos get pulled by the label. The save remains, an entry in your Watch Later that just says "Private Video" with no clue what it was.

The TikTok grid. Eight hundred saves, no titles, no descriptions, a grid of identical first-frame thumbnails. The recipe with the crispy potatoes is in there somewhere. You will not find it. The TikTok save button takes one tap, and finding anything in that grid later takes much longer.

The podcast minute you can't find. Episode 412. You loved a line about three-quarters of the way through. You forwarded the link to yourself. You did not write down the timestamp. Now you have to scrub through 87 minutes to find a 40-second moment, and the speaker phrased it slightly differently than you remembered.

The dead Twitter thread. A thread you saved 11 months ago. You go back. Account suspended. Or deleted. Or this one tweet removed. Twitter does not preserve a copy for you. The save is a 404.

The PDF that was a download page. What you really saved was a link to a download landing page that has since been reorganized. The PDF is somewhere on the same domain, maybe, under a different URL, or it's gone.

The Instagram post that went private. Or the account that got deleted. Or the reel the platform took down for music copyright. Same family of failure.

Most of these are silent. The save still sits in your list, looking like every other save. You only discover that it failed in the moment you actually need it, which is also the moment you can't fix it.

The internet does not preserve itself

This is not a fringe problem. Pew Research's 2024 link rot study looked at a representative sample of webpages over a decade. Their headline finding: 38% of webpages that existed in 2013 were no longer accessible in 2023. Of the tweets Pew followed for three months in 2023, nearly one in five were no longer publicly visible, because the account went private, was suspended or deleted, or the tweet itself was removed.

Nothing dramatic causes it. Sites rearrange. Companies shut down. Creators delete. Platforms change policy. Paywalls go up. Accounts get banned. Each one is a small private outage for whoever saved that link.

It does not take a decade, either. Of the pages Pew collected in 2021, about one in five were gone two years later. A save-list built over a few years keeps losing pieces, and nothing in the list shows which ones.

Link rot examples: how common it is

Link rot reaches places that are supposed to last. Three studies show the scale:

News, government and Wikipedia. In Pew's 2024 study, 23% of news webpages and 21% of government webpages had at least one broken link. On Wikipedia, 54% of pages had at least one link in their References section pointing to a page that no longer exists.
The New York Times. The Harvard team checked links in Times articles from 1996 to mid 2019. A quarter of the deep links were completely inaccessible: 72% of links from 1998 and 43% from 2008, against 6% from 2018. Of the links that still opened, 13% in a sample of 4,500 had drifted significantly from what the article cited.
Courts and law journals. A 2014 study in the Harvard Law Review Forum found that 50% of the URLs in U.S. Supreme Court opinions and more than 70% of the URLs in the law journals studied no longer produced the information originally cited.

Personal saves sit on the same web. Pinterest is a common case, covered in why Pinterest links are broken.

Why the URL stopped being the save

A link used to feel like enough. Fewer pages sat behind paywalls, and fewer of the things people saved were videos, reels and threads on platforms that can take them down.

Today the link is the least reliable part of a save. It points to a server you do not control, hosting content the publisher can change or remove, on a platform that can shift policy or shut down. Anything you want to remember has to live outside that chain.

Many saving tools still keep only the address. A browser bookmark is a URL. A Pinterest pin is an image with a link back to its source. A link pasted into a notes app is still a URL. Read-it-later apps go further and keep the article text, so it is worth checking what yours does with videos, screenshots and pages behind a paywall.

A save in 2026 has to mean: the content, not the address.

What "save the content" actually means

If a save is going to survive, it has to capture the thing that mattered at the moment of saving: the text of the article, the transcript of the episode, the picture itself.

None of this replaces the publisher's page. It means the part you wanted to remember stays with you even if that page changes or disappears.

Check what your save method keeps

Before trusting an important save, verify the return path:

Open the saved item from the place where you expect to retrieve it.
Check whether you see stored content or are simply redirected to the original URL.
Confirm that the title or saved context will still help you recognize it later.
For material you truly need, test the save again rather than assuming the bookmark itself is a copy.

The useful question is not only “Did I save it?” It is “What will still be here when I come back?”

How to tell if you have a save you can't open later

Three quick checks:

Open ten random items in your read-later app or browser bookmarks from more than a year ago. Count how many still load with the content you remembered.
Go to your YouTube Watch Later. Count the "[Deleted Video]" and "[Private Video]" entries. They were yours. They are gone.
Look at your TikTok saves. Pick a video you remember loving. Can you find it in under two minutes?

If those checks make you wince, look at where your saves live. Reading them more often will not bring back a deleted video.

Next time you save something that matters, do one extra thing. Save it somewhere that keeps the content, not the link. Screenshot the paragraph. Forward the page text. Use a tool that grabs the transcript. Make sure the thing that made you save it is the thing being saved.

How to prevent link rot in the things you save

Keep a copy of the part you need at the moment you save it. You cannot stop a site from deleting a page, but you can stop that from deleting your copy.

A page you may cite or quote. Paste its address into the Wayback Machine's Save Page Now, which captures "a web page as it appears now for use as a trusted citation in the future". It saves that one page, once, according to the Internet Archive's help. Keep the archived address next to the original.
A page you want on your own computer. Save a file: a saved page, a PDF or a full copy from an extension. Our guide to a Chrome extension to save web pages compares the options.
A video, thread or post. Write down or screenshot the part that mattered, with the title and the author. If the post disappears, you still have the line you saved it for.

For a link that is already dead, enter the address in the Wayback Machine and press Browse History to see whether a copy was captured before it broke.

Where dEssence fits

dEssence is built for the return trip. When you send it a page, Tracy works out what it is and keeps the useful details with the save, so you are not relying on the link alone. Later, you can ask for it in your own words instead of hoping the original link still works.

You can save through the Chrome extension, Telegram bot or web app at dessence.ai.

Link rot questions

What does link rot mean?

Link rot means a link no longer leads to the page it was made for. The page was deleted or moved, or the site is gone, so the link ends at an error or the wrong page.

What is the difference between link rot and content drift?

With link rot, the page is gone. With content drift, the link still opens but the content has changed, so it no longer shows what was originally saved or cited.

How common is link rot?

Common. Pew found that 25% of the pages it collected from 2013 through 2023 were no longer accessible in October 2023, and 38% of pages from 2013 were gone.

Can you fix a rotten link?

Sometimes. Enter the dead address in the Wayback Machine and check for an archived copy. For new saves, keep the content itself or an archived copy so you do not need the original page.