Issue The Card Or Move The Money - Pick Your Fintech Primitive

Ask a founder what "embedded finance" means and you will usually get a wave of the hand. It sounds like one thing. It is really two. There is the card - the plastic or the digital token a company puts in front of its users. And there is the money - the ACH files, the wires, the checks, the settlement that quietly happens behind the card. Highnote and Increase are what you get when two teams each decide to be great at exactly one of those halves.

Both were founded in 2020. Both landed on the Forbes Fintech 50. Both came out of the payments world with a grudge against how hard the incumbents made everything. And then they walked in opposite directions. Highnote went toward the card. Increase went toward the rails. Understanding that split is the fastest way to make sense of a category that otherwise reads like a wall of acronyms.

The card side: Highnote

Highnote was started by John MacIlwaine and Kin Kee, two executives who ran the machine at Braintree - the PayPal-owned processor that spent years trading punches with Stripe and Adyen. MacIlwaine had a long resume before that: 25 years across Visa, a stint as CIO at the prepaid issuer Green Dot, and CTO at LendingClub. Kee ran architecture. They knew where the bodies were buried in card issuing, and they built a platform to dig them all up at once.

The pitch is consolidation. Most companies that want to launch a card have to stitch together an issuer processor, a program manager, a ledger, fraud tooling, and - separately - an acquiring relationship if they also want to accept payments. Highnote's argument is that these should live on one API with one real-time ledger, so that fraud rules, spend controls, rewards, and reconciliation all read from the same source of truth. Issue debit, credit, or commercial cards; accept payments; keep the books - without four vendors disagreeing about what happened.

The market rewarded the bet. In January 2025 Highnote raised a $90 million Series B - later extended with existing investors - at a valuation north of $750 million, led by Adams Street Partners with Oak HC/FT, Costanoa, WestCap, and Pinegrove joining. That took total funding to roughly $155 million. The customer list grew past 70 enterprises, names like BNY Mellon, Netevia, and Mudflap among them, running small-business credit, payroll cards, and fleet programs. Highnote also cleared Visa's issuer-processor certification, the kind of unglamorous milestone that actually lets a card program exist.

The money side: Increase

Increase is the mirror image. Its founder, Darragh Buckley, was Stripe's first employee. For six years he did the least Instagrammable job in fintech: negotiating and scaling Stripe's bank relationship with Wells Fargo, and building the programmatic controls for deposit accounts, money movement, regulation, and privacy. He left to build the thing he had spent years wishing existed - a clean API that plugs straight into the plumbing of American money.

Where Highnote wants to help you issue a card, Increase wants to help you move a dollar. Its API covers ACH transfers, wires, real-time payments, and checks, connecting directly to the Federal Reserve's ACH network, Fedwire, and Visa, delivered through partnerships with FDIC-insured banks. The design philosophy is almost aggressively boring: money movement should be predictable, observable, and free of surprises. That is a feature, not a shrug.

The customers tell the story. Stripe, Ramp, and Gusto lean on Increase to move, store, and lend money at a scale measured in hundreds of billions a year. And unlike most of its peers, Increase did it while self-funded and profitable - the company reported turning a profit rather than chasing the next round. Then in 2025 it did something the rest of the category only talks about: it bought a bank. Buckley took 100% of the voting shares of Twin City Bancorp, a small Washington lender, with Federal Reserve approval landing in mid-2025. By July 2026 that became Increase Bank, an FDIC-member institution.

Same wave, opposite ends

This is where the comparison gets useful. People want a winner. There isn't one, because they are not really running the same race. Here is the split, laid out plainly.

The lines do blur at the edges. Increase also offers card issuing; Highnote also moves money inside its ledger. But their centers of gravity sit at opposite ends of the stack. Highnote is the front end you hand to a user. Increase is the back end that quietly clears the payment. A lot of companies eventually need both - which is exactly why treating them as rivals misses the point.

The lineage tells you a lot

It is worth sitting with where these founders came from, because the pedigree shaped the product. Braintree and Stripe were the two poles of a generation of payments engineering - Braintree the older, PayPal-absorbed processor, Stripe the developer-first upstart that ate the market. Highnote carries Braintree's DNA: deep card expertise, a bias toward owning the full processing stack, a comfort with the messy realities of issuer certification and program management. Increase carries Stripe's: obsessive API design, documentation you can actually read, and a belief that infrastructure should feel like a well-behaved library rather than a bank teller.

That difference shows up in how they talk about themselves. Highnote sells a "single platform for modern payments" - the language of consolidation, of replacing a tangle of vendors. Increase describes a bank "built by a team of product-obsessed operators for ambitious companies." One is selling you fewer moving parts; the other is selling you rails that behave. Both are reactions to the same frustration, filtered through two different engineering cultures.

There is also a funding tell. Highnote took the venture path - roughly $155 million across seed, Series A, and a $750 million-plus Series B - because building a card platform with acquiring and credit baked in is capital-hungry, and speed matters when Marqeta and Stripe Issuing are in the same market. Increase went the other way and stayed self-funded and profitable, which is rare enough in fintech to be a statement of identity. Owning your own bank charter is not something you do to impress investors; it is something you do because you have decided to control your own destiny at the root of the stack.

What you can actually steal from this

If you are building anything that touches money, the lesson is not "pick Highnote or Increase." It is: name your primitive before you shop. The single most expensive mistake in embedded finance is buying a card platform when you needed money movement, or wiring up bank rails when what your users actually wanted was a branded card in their wallet. The two companies are, in effect, a decision tree drawn in public.

There is a second, quieter lesson in how they were built. Both founders came from inside the giants - one from Stripe, one from Braintree - and both concluded the incumbents had made a solvable problem too hard. Increase answered by making money movement boring and owning its own bank. Highnote answered by collapsing five vendors into one ledger. Different tactics, identical instinct: remove the layer that adds risk without adding value. That instinct is portable to almost any industry with a bloated middle.

And notice what neither of them did. Neither tried to be everything. In a market where every pitch deck promises the full stack, two of the more respected infrastructure companies got there by picking one primitive and refusing to blur it. The clarity is the product.