Nothing automatic happens. There is no beneficiary field on a blockchain, no next-of-kin process, and no institution to notify. If nobody can find and use your keys, the coins stay where they are — visible on the chain, permanently unspendable.
That is the whole problem, and it is a design consequence rather than an oversight. Self-custody means no third party can move your funds. Your heirs are a third party.
Two situations, two different failures
Crypto on an exchange is a claim against a company, so it behaves like any other account. Most large exchanges have a deceased-account process requiring a death certificate and grant of probate or its local equivalent. It is slow — months is normal — and it depends on your executor knowing the account exists. The asset is recoverable; the obstacle is administrative.
Self-custodied crypto has no process at all. Whoever holds the recovery phrase controls the funds, full stop. If your heirs cannot find it, cannot read it, or do not understand what it is, the money is gone permanently — and no court order changes that, because there is nobody for the court to order.
The failure is almost never “the heirs were locked out by cryptography”. It is that they never knew a wallet existed, or found a card of words and threw it away.
The mistake that ruins everything: putting the phrase in your will
This seems like the obvious solution and it is close to the worst one.
In many jurisdictions a will becomes a public record once it goes through probate. Anyone can request a copy. A recovery phrase written into a will is therefore a recovery phrase published to the world — and the funds will be gone long before the estate is settled.
Even where a will stays private, it passes through solicitors, executors and court staff, any of whom could copy it, and it is typically drafted years before it is used, so the phrase in it may be stale.
The correct pattern separates the two: the will says the asset exists and who inherits it. A separate, private document says how to access it. The will points; it does not contain.
What a workable plan looks like
- An inventory. What exists, roughly how much, and where — exchanges, wallets, devices. Without this, nothing else matters, because nobody looks for what they do not know about.
- A named person who knows there is something to find. Not the details; just that a plan exists and where the instructions are. A perfect plan nobody knows about is no plan.
- Access instructions, written for someone who does not know what a wallet is. This is its own job and covered separately.
- The secret itself, stored so that it survives you without being available to anyone while you are alive — a sealed envelope with a solicitor, a bank box your executor can reach, or split across trustees.
- A legal wrapper. The will should mention the asset and name who inherits it, referring to the instructions without reproducing them.
The trade-off you cannot escape
Every inheritance mechanism moves along one line: the easier you make it for your heirs, the easier you make it for a thief.
A phrase in a sealed envelope with a trusted relative is simple and works — and that relative can take the funds today. A 2-of-3 arrangement where no single person can act alone is far safer and demands more of everyone involved. There is no arrangement that is both maximally secure and effortlessly inheritable, and anyone selling you one has hidden the trade rather than solved it.
Reasonable options, in rising order of complexity:
- Sealed instructions with a solicitor, released on death. Simple, legally familiar, and dependent on that firm’s integrity and continued existence.
- Shamir backup (SLIP-39), splitting the phrase so that, say, any 2 of 3 shares restore it and one reveals nothing. Supported natively by some hardware wallets.
- Multisig, where separate keys held by different people are needed to spend. The strongest option and the most demanding.
Test it, because an untested plan is a guess
The check that matters: could the person you have chosen actually do this, with only what you have left behind, without you in the room?
The way to find out is a dry run — hand them the instructions for a wallet holding a trivial amount and watch them try. It is uncomfortable and it is the only thing that converts an intention into a plan. Most first attempts fail on something mundane: a word nobody can read, a step that assumed knowledge, a device with a PIN nobody has.
And review it when things change — a new wallet, a new device, a passphrase added, a relationship changed. A plan describing a setup you abandoned three years ago will fail exactly as completely as no plan at all.