What actually "rots"
The link itself — the URL string — doesn't change or decay in any technical sense. What rots is the destination: a page gets deleted during a site redesign, a company shuts down and its domain lapses, or content gets reorganized under a new URL structure without anyone setting up a redirect from the old one. The link keeps pointing to the same address; that address just stops leading anywhere useful.
Why it happens so often
Websites are not static documents — they get redesigned, migrated to new platforms, restructured, and eventually retired, and none of that process is obligated to preserve every old URL. A blog might move from one URL pattern to another during a platform switch; a business might close and let its domain expire; a page might simply get deleted because someone decided it was no longer needed. None of these are unusual or negligent events — they're just the normal lifecycle of content on the web, and every one of them can break a link pointing at the old address.
Where it causes the most real damage
Link rot is most costly in places meant to last: academic papers and citations, legal documents, news articles referencing sources, old blog posts recommending tools or resources, and internal documentation linking out to vendor pages. A citation or reference that pointed to solid supporting material when it was written can become a dead end years later, undermining the credibility of the document that relied on it — with no way for a reader to know what used to be there.
Does a URL shortener cause link rot, or help with it?
These are two separate questions people often conflate. A shortener's own link staying alive is entirely within the shortener's control — as long as the service is running, the short URL itself keeps resolving. What it can't control is whether the destination it redirects to still exists; if that page gets taken down, the short link will faithfully redirect to a dead page, the same as any other link would. The difference is what happens next.
Why a dynamic link is more resilient to this than a static one
If a destination goes offline, a short link you control can be updated to redirect somewhere else — an archived version of the page, a replacement resource, or even a short explanation of what happened — without needing to track down and edit every place the original long URL was ever shared. A raw, hardcoded URL pasted directly into a document or an old blog post has no such recovery path; once its destination is gone, every copy of that link is permanently broken with no way to fix it centrally.
Practical habits that limit the damage
For content you expect to have a long shelf life — reference material, evergreen guides, anything likely to be cited elsewhere — favor short links you control over pasting in raw destination URLs directly, specifically so you retain the ability to redirect them later if needed. For your own site's internal restructuring, setting up a 301 redirect from an old URL to its new location prevents your own past links (and anyone else's) from rotting the moment you reorganize.
Periodically checking what you've already shared
Link rot compounds quietly — a resource page with fifty links accumulates dead ones a little at a time, and nobody notices until a reader reports it or you happen to click through yourself. Revisiting older content periodically, especially anything positioned as a reference or resource list, and checking whether the links still resolve is one of the simpler maintenance habits that meaningfully improves the long-term reliability of anything you publish.
It is not a reason to avoid linking at all
The existence of link rot is sometimes used as an argument for avoiding external links altogether, but that trade-off usually isn't worth it — a working, useful citation today is valuable even if it might need updating in a few years. The more practical response is building in the ability to fix links later rather than avoiding the risk by not linking to anything external in the first place.
How link rot affects SEO, and why internal rot is the worse problem
Search engines generally treat a page riddled with dead outbound links as a weaker signal of quality and upkeep, since it suggests the page hasn't been maintained. But internal link rot — links pointing from one of your own pages to another of your own pages that no longer exists — tends to do more concrete damage. A dead internal link wastes a search engine's crawl budget on a URL that returns nothing useful, and it can orphan pages that were only discoverable through that link, cutting them off from the rest of your site's link structure entirely. Restructuring a site without setting up redirects for the old URLs is one of the more common, avoidable ways this happens.
A simple mental model
It helps to think of every link as making an implicit promise about the future: "this will still be here." Most of the time that promise is kept, but the web has no built-in enforcement mechanism for it, so over a long enough timeline a meaningful share of links will eventually break. The practical takeaway isn't to treat every link with suspicion — it's to recognize that anything meant to last should be built with the assumption that at least some of its links will eventually need fixing, and to make that fixing process as low-friction as possible.