If you self-custody Bitcoin and you’ve thought about what happens to it when you die, you’ve probably hit the same wall I did: the existing options all suck.

There are four-ish, broadly:

Give your heir the seed. They lose it, spend it before you die, family drama erupts, you change your mind and can’t put the toothpaste back in the tube. Bad option.

Multisig with family. Now your inheritance depends on family members understanding key management years from now. Your kid’s seed gets reset when their phone breaks. Your spouse loses interest after the third “what’s a Bitcoin.” Multisig is fundamentally a coordination problem, and inheritance is the worst possible time to have one.

Custodial inheritance service. Casa Covenant, Unchained, Nunchuk Inheritance — these work, but they require you to put your Bitcoin into a model where a company holds a key. Companies fail. Companies get hacked. Companies decide they don’t want your kind of customer anymore. You’re back to “trust someone else,” which is the exact thing self-custody exists to avoid.

Write it in your will. The will says “leave my Bitcoin to my daughter.” Your Bitcoin is in cold storage controlled by a seed nobody knows. The will is a directive, not a transfer. Without something underneath it, that paragraph is the world’s saddest legal fiction.

I went through these in order over a few months. Each one made me a little more annoyed. The whole reason I self-custody is that I don’t want to trust anyone — and every Bitcoin inheritance product asked me to put that trust back in.

It turns out Bitcoin has had the right primitive for this since 2013.

nLockTime, finally useful

nLockTime is a field on every Bitcoin transaction. Set it to a future block height and the network refuses to include the transaction until that block is mined. The cryptography is real now; the broadcast eligibility is gated by the chain itself.

That’s enough to build inheritance without trusting anyone. You sign a transaction today that pays your heir, with nLockTime set to ~4 years out. You give them the signed bytes. They sit on them. When the unlock block fires, they broadcast. Funds settle. The Bitcoin network does the enforcement — the same chain that does anti-double-spend does anti-broadcast-early.

No company. No multisig coordinator. No custodial relationship. The trust model is “Bitcoin works.”

I built Bitheritance because nobody else seemed to be building this. The protocol primitive has been sitting in Bitcoin for over a decade, more or less unused for end-user inheritance, while well-funded companies built around the much harder problem of safe custody.

Constraints I refuse to compromise on

When I started, I picked four things I’d never trade away:

The server holds no keys, ever. The WASM that builds and signs your transaction runs in your browser. Our backend learns the output (signed hex), not the input (your seed). This is enforced by where the code runs, not by me promising to behave.

The artifact works without us. The signed transaction is a standard Bitcoin transaction. Sparrow, Electrum, Bitcoin Core CLI, mempool.space’s broadcaster — any of them will broadcast it. If Bitheritance shuts down tomorrow, your heir’s inheritance still works.

The heir might never have opened a Bitcoin wallet. The handoff kit is a printable PDF with step-by-step instructions for someone who’s never done this. The technical literacy gap is the operational problem nobody else is taking seriously.

The unlock might be 30 years away. Recipient address can’t change. Fee is baked in at sign time. Post-quantum is a real concern past the 2030s. Bitheritance doesn’t solve these — it’s honest about them in the wizard.

What’s still hard

I’m not pretending this is finished. Real things that keep me up:

  • Fees in 30 years. We offer an optional fee-bump anchor for CPFP, but it’s not magic.
  • The heir handoff. The signed transaction is just bytes; getting it from your hands to your heir’s in a way that survives 30 years is mostly an operational and social problem, not a technical one.
  • Post-quantum. Long plans are bets on the protocol situation staying favorable.
  • Pricing the watchtower. Polling chain state for users for years before they need anything costs real money. We charge $2/mo for it (signet is free always).

The trust model is solved. The operational stuff is mostly not. That gap is where the next few years of this project are going.

If you’ve thought about Bitcoin inheritance and bounced off the existing options — this is the one I think you should actually try. Signet is free; do that first.