Link rot: what happens when a shortener shuts down
Shortening a URL replaces a name you control with a name someone else controls. That is the whole trade, and it is usually worth it. But it makes a short link a dependency, and dependencies deserve the same scrutiny as any other.
What actually breaks
When a shortener disappears, the destination pages are all still there. What is lost is the mapping, and the mapping lives in places you cannot edit.
There is a second-order risk that is less obvious and worse. When a shortener's domain lapses, someone else can register it, and every link ever created on it now points wherever the new owner chooses. Your old post now recommends something you have never heard of. This has happened, repeatedly, and it is the strongest argument against the cheapest possible provider.
This is not hypothetical
The shortener graveyard is large. Services run by major platforms have been retired with a year's notice; independent ones have closed with a month, or none. Some kept redirects alive indefinitely as a courtesy, and that courtesy has no expiry date attached to it, which is its own kind of uncertainty.
The pattern worth noticing is that free services die most often. A redirect table costs almost nothing to run but is not free, and a product with no revenue attached to it is the first line item cut when priorities change. Paying a small amount is not a guarantee of survival, but it does mean the service has a reason to exist.
Four habits that make it survivable
c1k.me/pricing still tells a reader what it meant. c1k.me/x7Fq2 tells them nothing.What c1k.me commits to, and what it cannot
Nobody running a small service can honestly promise to be here in ten years, and a page claiming otherwise should make you suspicious. What can be written down are the mechanics, and those are in the terms rather than in a blog post.
The deliberate design choice underneath these is that leaving stays cheap. A service that makes export easy has to keep earning the renewal, which is a healthier arrangement than one that relies on the cost of migrating.
Use shorteners deliberately
None of this argues against short links. It argues for knowing which category each one is in. A link in a chat message that will be irrelevant next week carries no risk worth thinking about. A link etched into product packaging is infrastructure and should be treated as such.
Most teams never make that distinction and use one mechanism for both. Making it takes about five minutes and removes the only genuinely painful failure mode.