B2B stablecoin payments: what is actually moving

B2B stablecoin payments are no longer a projection. The most careful public measurement, by Artemis with Castle Island Ventures and Dragonfly, drawing on transaction data from around 20 payment companies, found business-to-business stablecoin volume above $3 billion a month by 2025, roughly 30 times the level of early 2023. The same research put total stablecoin payment volume at about $390 billion a year, with B2B flows making up around 60% of it. Those figures cover only the firms that shared data, so the real market is larger.

This page sets out what is verifiably moving, where, and for what, and then the part the headline numbers tend to hide. Every figure below is public and sourced; where the data cannot support a claim, the claim is not made.

The numbers that hold up

Three data points anchor the picture. The Artemis research gives the B2B baseline: $3 billion-plus a month, concentrated in corridors linking the United States, China, and Hong Kong, alongside routes into Latin America, Africa, and Southeast Asia where correspondent banking coverage has thinned. Visa gives the card-network signal: its stablecoin settlement pilot reached a $7 billion annualized run rate in April 2026, up 50% quarter on quarter, and expanded to nine chains supporting more than 130 stablecoin-linked card programs. And the official sector supplies the context that explains the demand: the FSB concluded in October 2025 that the G20’s end-2027 targets for faster, cheaper cross-border payments are unlikely to be met at the global level.

Growth of this shape, from a small base, in specific corridors, while the incumbent improvement program slips, is what a working alternative looks like in its early phase. It is not yet what displacement looks like: $390 billion a year of stablecoin payments is still a rounding error against interbank flows, where a single day on CHIPS clears about $2 trillion.

Why 2025 was the turn

The volume curve bent when the rulebook arrived. The GENIUS Act, signed in July 2025, gave the United States a federal framework for payment stablecoins, with the OCC publishing detailed implementing proposals in February 2026. MiCA’s stablecoin provisions reached full enforcement in the EU on 1 July 2026, sorting the market into authorized e-money tokens and everything else. Hong Kong’s Stablecoin Ordinance came into force in August 2025. Businesses that had watched from the sidelines were not waiting for the technology; they were waiting for a compliance officer to be able to say yes.

The supply side answered in kind. Card networks moved stablecoin settlement from experiment to program. Banks began launching deposit tokens and routing regulated coins for clients rather than ceding the flow. And the acquisition prices paid for stablecoin infrastructure, Stripe’s reported $1.1 billion for Bridge and Mastercard’s reported takeover of BVNK, signaled that the incumbents expect these rails to carry meaningful volume. For a corporate treasurer, each of these lowered the career risk of routing real payments over the new rail, and career risk, more than latency, is what had held B2B volume back.

The use cases that are live

Supplier payments in hard corridors. The clearest product-market fit is paying suppliers where the correspondent chain is slow, expensive, or receding. An importer paying a manufacturer across a de-risked corridor can move value in minutes rather than days, without pre-funding accounts along the route.

Intercompany treasury. Groups with subsidiaries in many countries use stablecoins to reposition cash between their own entities, especially into and out of markets where repatriation through banks is slow. The transfer is between wallets the group controls, which simplifies counterparty risk while leaving the compliance obligations intact.

Payout networks. Marketplaces, payroll providers, and gig platforms paying thousands of contractors or sellers across dozens of countries use stablecoins as the distribution leg, converting to local currency at the edge. This is the use case where volume compounds fastest, because every new market adds recipients without adding intermediary banks.

FX settlement. Trading firms and payment companies settle the currency leg of FX trades in stablecoins to compress settlement windows and reduce settlement risk between counterparties, particularly outside banking hours when traditional payment-versus-payment infrastructure is closed.

What unites the four is the shape of the corridor, not the size of the company. Stablecoin B2B volume concentrates where the correspondent chain is at its worst: multiple intermediaries, thin liquidity, restrictive cut-offs, or outright de-risking. Where domestic instant rails or dense correspondent coverage already serve a route well, the data shows businesses staying put, which is exactly what a rational treasurer would do.

How to read the data

Stablecoin volume numbers deserve more skepticism than they usually get, in both directions. Raw chain-level totals wildly overstate payment activity, because most on-chain transfer volume is trading, market-making, and internal rebalancing rather than anyone paying anyone. That is what makes the Artemis approach the reference point: it builds from the ground up, counting only flows reported by named payment companies, which is why its totals look small next to headline on-chain figures. The cost of that rigor runs the other way. Twenty-odd reporting firms cannot see the whole market, so the $3 billion monthly B2B figure is a floor, not a ceiling, and it excludes the bank-run and card-network flows measured separately.

Two more reading rules help. First, distinguish settlement volume from payment volume: Visa’s $7 billion run rate is the network settling its own obligations with partners in stablecoins, a different (and in some ways stronger) signal than businesses paying invoices. Second, watch corridor concentration. Growth clustered in a handful of corridors is evidence of product-market fit in those corridors, not of a general migration; the corridors where correspondent banking works well show far less movement.

What the averages hide

The friction has not disappeared; it has moved to the edges. A stablecoin transfer itself settles in minutes with finality, but most B2B payments still begin and end in bank money, and the conversion legs reintroduce what the middle removed: fees, cut-off times, and counterparty onboarding. Where both edges are thin, the all-in cost of the stablecoin route can approach the wire it replaced.

Compliance travels with the payment. Stablecoin transfers between institutions carry the same sanctions, AML, and Travel Rule obligations as wires, and the businesses moving real volume are the ones that treated that as a design input rather than an afterthought.

And the market is fragmented. Volume splits across coins with different regulatory standing (USDT dominates emerging-market corridors, USDC the regulated ones), across chains with different costs and finality behavior, and across providers with different licenses and coverage. A business that integrates one provider inherits its limits; one that integrates several inherits a reconciliation problem.

What to watch next

Four indicators will show whether the curve keeps bending. The pace of GENIUS Act licensing once the OCC’s rules are final, because each permitted issuer widens the set of coins a US compliance team can approve. Coin substitution inside EEA corridors, as MiCA enforcement pushes volume from unauthorized tokens toward authorized e-money tokens; the aggregate numbers will hide this, corridor-level data will not. Bank participation, particularly whether deposit-token settlement reaches volumes worth reporting, which would mark the moment tokenized bank money starts competing for the same flows. And the official sector’s own deadline: as the G20’s end-2027 targets approach with the FSB already calling them unlikely, the gap between mandated ambition and delivered improvement is the space every alternative rail grows into.

None of these depends on any single coin, chain, or network winning. That is worth noticing, because the businesses moving volume today are making the same observation.

Where Frame fits

Frame is the settlement layer for global finance: one integration to orchestrate payments at scale across fiat rails, stablecoins, and tokenized deposits, with compliance enforced on every transaction and settlement in seconds instead of days.

The B2B stablecoin numbers above are evidence that the settlement leg of cross-border payments is being rebuilt. Frame’s position is that no single coin, chain, or network wins every corridor, so the durable capability is the layer that routes across them. A payment enters through one integration; Frame routes it over whichever rail fits the corridor and the policy that governs it, and its Rules Engine evaluates every transaction against those policies inside settlement, so a transfer that cannot satisfy them does not settle. For banks, payment providers, exchanges, platforms, and enterprises, that turns stablecoins from a separate integration project into one more rail the same layer already serves.