Five Ways to Hold Bitcoin Without a Ledger, and the Bill Attached to Each

Ledger spent a decade selling the idea that a small USB device was the final word in bitcoin custody, and the pitch held up right until the supply chain got involved.

The alternatives are real now. Spot ETFs, passkey logins, third-party multisig accounts, each pitched as the safer, easier, more grown-up option. Every one of them fixes something the hardware wallet got wrong and introduces a new failure point that the marketing deck skips past. That's the trade nobody prices honestly, so let's price it.

The Sequence

Start with the wound that built the whole category. February 2014, Mt. Gox. Roughly 850,000 bitcoin gone from what was then the largest exchange on earth, about 750,000 of it belonging to customers. The lesson hardened into doctrine fast. Not your keys, not your coins.

So the industry sold keys. Ledger and Trezor turned cold storage into a consumer product, and for years it was genuinely the best answer available. Plug in, confirm on the screen, sleep at night.

December 2020. Ledger's customer database leaked. 272,000 names, emails, phone numbers, and in some cases home addresses. No coins moved that day. The damage was slower and uglier, phishing campaigns, threats, people getting visited at their front doors.

December 14, 2023. The Ledger Connect Kit library, a chunk of code embedded in dozens of decentralized apps, got compromised. Around $600,000 drained out of user wallets in roughly two hours before the malicious payload was pulled. Read that again. A hardware wallet vendor's software dependency became the attack surface for people who thought they'd removed the attack surface.

January 10, 2024. The SEC approved 11 spot bitcoin ETFs in a single afternoon. Custody went institutional, and the self-custody crowd suddenly had a competitor with a marketing budget and a brokerage login.

February 21, 2025. Bybit, roughly $1.46 billion in ether moved out through a single traced transaction trail. The largest crypto theft on record. Custodial, sure, but a useful reminder that custody debates aren't academic.

And running underneath all of it, from 2022 into 2025, passkeys. Apple, Google, and Microsoft shipped them across phones and browsers. Exchanges started bolting them onto login flows. The seed phrase began to look optional for the first time.

What Each Trade Costs

Here's where the honesty usually stops. Take the ETF. IBIT charges 0.25%, GBTC still charges 1.5%, and neither number tells you what you actually bought. You bought a share of a trust. The bitcoin sits with a custodian, and that custodian's hot wallet is the thing that has to hold. The marketing says you own bitcoin. The share register says otherwise.

Is that a bad product? No. It's a fine product for a retirement account that can't touch a private key. But call it what it's. It's bitcoin exposure wrapped in a regulated wrapper, and the wrapper has a fee, a counterparty, and a terms-of-service page that most buyers never open.

Passkeys are the more interesting shift, because they move the weak point rather than remove it. Your biometrics stay on the device, that part's real. But the recovery flow runs through an account recovery desk at a company with a support queue and a policy. Account recovery is the softest target in all of security, and always has been. So who's actually holding your coins now?

Then there's collaborative multisig, a 2-of-3 or 3-of-5 setup where a provider holds one key and you hold the rest. Casa runs around $10 a month, $120 a year, and it's the closest thing to an honest answer in this entire discussion. It keeps you in the loop, it removes the single point of failure, and it makes inheritance survivable. But here's the gap. The typical failure isn't a compromised key, it's a human being who set the thing up in 2021 and can't remember which hardware device pairs with which backup. Show me the audit on the recovery experience, not the cryptography.

Two things I'll say plainly. First, an ETF is a bitcoin tracking product, and the industry should stop letting it be marketed as ownership. Second, if the recovery flow is where every one of these alternatives breaks, why does every vendor bury it on page nine of a FAQ?

The Test Coming

Three concrete things to watch. ETF options cleared in October 2024, and every new options market pulls more coins deeper into custodian vaults and further from self-custody. Watch the share of total bitcoin supply sitting in ETF trusts. That number is the scoreboard, and it only moves one direction while fees stay low.

Second, passkey adoption on exchanges. If the big venues make passkeys the default login by the end of 2026 and pair that with a genuine, tested recovery path, you'll see seed phrases start to disappear from consumer onboarding entirely. That's a real change, and it hands a huge amount of practical control to three or four technology companies.

Third, quantum migration. NIST finalized its post-quantum standards in August 2024, ML-KEM, ML-DSA, SLH-DSA, and the working estimates for moving bitcoin's signature scheme run somewhere between five and ten years of coordinated consensus work. Nobody's doing that work at the pace the timeline demands.

So let's apply the standard the industry set for itself. Self-custody was supposed to mean removing trust from the equation. Every alternative on the table right now reintroduces it somewhere else, and the sellers know exactly where. The burden of proof sits with the team, not the community. Ask each vendor where the recovery breaks, then ask to see the test results. Skepticism isn't pessimism. It's due diligence.