It was almost never the technology
The contracts worked. That is the first thing the record makes obvious and the thing most retrospectives get wrong. 6 of 6 recorded contracts are still live, still holding their supply, still answering calls. Not one of these projects ended because the chain failed or the code broke.
They ended for the reasons anything inside a large company ends: the person who championed it moved on, the budget line was cut, the metrics never justified the headcount, or the legal position got uncomfortable. Ordinary corporate causes of death, applied to an unusually permanent object.
The five patterns
1. The campaign mistaken for a product
Most of these were marketing campaigns with a launch date and no operating plan. A campaign is allowed to end — that is what campaigns do. The problem is that a campaign which mints tokens leaves something behind that does not end, and nobody had budgeted for the afterwards.
2. The champion leaves
Almost every one traces back to a small number of internal advocates. Where the record shows a shutdown following a leadership change or a restructuring, it is usually not that the new leadership was hostile — it is that nobody was left whose job it was to defend it.
3. Regulatory temperature
Some closures cite regulatory uncertainty directly. In at least one case a regulator did more than create uncertainty: a members' club that sold roughly $14.8m of tokens as restaurant memberships settled with the SEC, paid $750,000, and unwound the token side of the business entirely. Notably, two commissioners publicly disagreed with their own agency about it — which tells you how unsettled the ground was.
4. No reason to come back
A token that unlocks a one-off drop has done its job the moment the drop happens. Several of these had no recurring event, no renewing benefit and no reason for a holder to check in twice. Engagement did not decline so much as complete.
5. The quiet ending
The most common ending is no ending. No announcement, no wind-down post, no note to holders. The dedicated subdomain simply stops resolving one day, and the only way to date it is to notice the DNS record has gone. Two entries here ended exactly this way.
What happens next is the interesting part
Shutting down a normal product ends it. Shutting down a tokenised one does not, and this is the gap nobody planned for. The company stops. The contract carries on, because a contract has no off switch and its issuer has no standing to reach into anyone's wallet.
The clearest case on the record: Nike RTFKT / CloneX. The operator behind it stopped, and in the last 14 days there were 263 transfers anyway — more activity than most of the projects here that are nominally still alive.
So a brand that exits leaves a thing in the world carrying its name, which it no longer operates, cannot recall, and does not support. Every one of these is now an orphan asset with a logo on it. Nobody wrote that down as a risk in the launch deck.
Figures measured directly from a public node — see how, or take the raw data.
What the survivors have in common
The sample is small and honesty requires saying so — this record has 1 clear survivor against a much larger number of endings. But the pattern is consistent with what you would expect, and it is unglamorous:
The token was attached to something the operator was going to keep doing regardless.
A conference that happens every year. A restaurant that serves dinner every night. Where the underlying activity would exist with or without the token, the token has a job to keep doing. Where the token was the activity, it ran out of reasons to exist at roughly the speed of the marketing budget.
Which is a fairly deflating conclusion: the successful ones are successful because of the ordinary business underneath, and the token is a membership card. That is not a failure of the idea. It is just a much smaller idea than the launch posts implied.
If you are doing this anyway
People still launch these, and some should. Three questions the record suggests are worth answering before you do, in writing, in the launch document:
- What does a holder get in year three? If the honest answer is nothing, you have a campaign. Run it as one, say so, and do not mint something permanent to commemorate a temporary thing.
- Who owns this when the champion leaves? Not aspirationally — which team, in which budget. Every entry here had an internal advocate. Every ending followed that person's departure or their budget's.
- What is the wind-down plan? What you tell holders, what happens to the metadata, who keeps the domain alive, whether the contract is left in a defined state. Not one project on this record appears to have published one before it was needed. Writing it at launch costs an afternoon; not having it is why several of these entries have no death date at all.
Answer those three and you will not appear here — or if you do, it will be as an entry that ended on purpose, which no one has managed yet.