Hardware Wallets in 2026: Touchscreens, NFC Cards, and the $500M Market Nobody Wants to Fund
Six form factors, a dozen vendors, and roughly half a billion dollars in annual device revenue. Here's who's actually making money in hardware wallets, who's burning cash on firmware they'll support for a decade, and what your threat model should cost you in 2026.
Do you actually need a hardware wallet in 2026? Short answer, yes, if you're holding anything you'd be annoyed to lose. The longer answer is that the device you buy matters a lot less than the recovery plan you'll probably never test until the day you need it.
So let's talk about the market behind the metal and plastic. Because the form factor war going on right now says more about the business than it does about security.
What the Devices Actually Cost
Start with the raw numbers, because they tell you where the money is. Ledger's Stax runs about $279 and the Flex sits around $249. Both have touchscreens. Trezor's Safe 5 is $169, the Safe 3 is $79. Coldcard's Mk4 is $158 and it doesn't even have a battery, you feed it PSBTs on a microSD card. Keystone 3 Pro is $129 and communicates purely through QR codes, no USB, no Bluetooth, no cable. BitBox02 lands near $150 out of Switzerland. Foundation's Passport runs $199 and only does Bitcoin. Tangem sells NFC cards, roughly $55 for a three-pack, and the card is the wallet.
Six form factors. Six different bets on what paranoid looks like.
Ledger sold more than 6 million devices since 2014. Trezor crossed 2 million back in 2021, having shipped its first unit in 2014. Those two control the bulk of a market that generates somewhere between $400 million and $600 million a year in hardware revenue, depending on whose estimate you trust. That sounds like real money until you remember how much of it goes back out the door.
Burn rate tells you more than valuation. A hardware company carries inventory, tooling costs, secure element procurement, certification fees, and then a decade of firmware support for every unit it ever ships. That last line item is the killer. You sell a $79 device once and you owe that customer security patches until they die.
Why Form Factor Became the Whole Fight
Hardware wallets used to be one thing. A little screen, two buttons, a USB cable. That was the entire product category for most of a decade.
Then FTX collapsed in November 2022 and self-custody stopped being a niche hobby. Ledger's device sales spiked. Trezor's did too. Everyone suddenly wanted their keys off an exchange.
And then in May 2023 Ledger announced a seed recovery service and the same customers who'd just bought in turned on them hard. The idea was simple. Split your seed phrase into encrypted shards and let three custodians hold pieces so you could recover your funds if you lost your paper backup. Opt-in. Voluntary. It still triggered a revolt because the whole point of a hardware wallet is that nobody else can touch your keys. Ledger walked it back to a beta, then eventually relaunched it as a paid subscription around $9.99 a month.
There's also the 2020 breach. Ledger's customer database leaked, exposing around 272,000 email addresses and 9,500 phone numbers. People got phishing texts and, in some cases, threats at their homes. That one still shapes how the industry talks about privacy.
Here's the takeaway. The last four years pushed buyers away from brand loyalty and toward threat models. Do you want a screen or do you want an air gap? Do you trust Bluetooth firmware or do you want QR codes and an SD card? The vendors didn't segment this market on purpose. The market segmented itself after being burned.
What the People Writing the Checks Say
Sources close to the deal say the venture math on hardware wallets has gotten awkward. Ledger's June 2023 round, 100 million euros, valued the company at roughly 1.3 billion euros and it was led by 10T Holdings. That's the headline. The round valued the company at a number that assumed recurring revenue, and recurring revenue is exactly what hardware has never reliably produced.
The check writers are getting pickier. And you can see why. Passkeys are shipping in every major browser. Apple and Google keep hardening secure enclaves. Exchange custody has improved a lot since 2022, even if you don't love it. Meanwhile the hardware side competes on industrial design, which is a terrible place to compete because Apple already won that fight and your $279 touchscreen is never going to look like an iPhone.
Follow the cap table and you'll notice something else. Very few new hardware wallet companies have raised a Series A in the last 18 months. The category is funded, mostly, by incumbents with cash flow and a handful of Bitcoin-only shops that run lean on purpose. Coldcard's parent, Coinkite, has never taken outside money at scale. That's not an accident. It's a business model choice.
So who wins? My read is that the Bitcoin-only crowd has the better unit economics and the multichain crowd has the bigger addressable market, and neither of them is going to escape the firmware liability problem. If you're a founder pitching me a hardware wallet right now, I want to know what your average support ticket costs you and how many firmware engineers you'll need in five years. Not your security architecture. Everyone's architecture is fine.
What to Watch in the Next Two Quarters
Three things, concretely.
First, post-quantum firmware. NIST finalized its first post-quantum standards in August 2024 and wallet vendors have been slow-walking the migration. Watch which companies ship signed firmware updates with hybrid post-quantum key schemes first. Whoever gets there early owns the enterprise and custody-institutional conversation.
Second, NFC. Tangem built an entire company on a card you tap, and the form factor maps cleanly onto payments. If a major issuer or neobank bundles a self-custody card with a checking account in 2026, that's the moment hardware stops being a crypto product and starts being a financial product. That's the real catalyst to watch, not another touchscreen.
Third, consolidation. There are too many vendors for a market this size. I'd put decent odds on at least one acquisition in the next two quarters, probably a Bitcoin-only shop getting absorbed by a custody provider that wants a hardware arm. Watch the announcements around the big spring conferences.
And one more thing worth saying out loud. Don't spend $279 on a touchscreen you'll use twice. Buy the $79 device, buy two of them, write your seed down properly, and actually test a recovery before you've got real money on it. That's the whole game. The rest is industrial design.