Pay by bank has been the next big thing in payments for about a decade, which is usually a sign that the story is more complicated than the pitch. After our conversation with Johannes Kolbeinsson on Fintech Garden, one framing stuck with us: in most markets, pay by bank is not primarily coming for cards. It is coming for cash and for clunky manual bank transfers. That distinction changes what payment leaders should actually prepare for.
Where account-to-account wins today
Account-to-account payments shine where cards were never strong: large-value transactions where interchange hurts, bill payments, invoice settlement, and markets where card penetration stayed low while banking apps became universal. In those niches, the value proposition is direct: lower cost, instant settlement, no chargebacks in the card sense, and a checkout that lives inside a banking app people already trust.
Where cards remain strong, in everyday consumer commerce with rewards, familiar dispute rights and frictionless credentials on file, displacement is slow. Consumers do not abandon a payment method that works. They adopt new ones where the old ones were never good.
The operational reality behind the checkout button
For a payment company, offering pay by bank is not one integration, it is a portfolio of them. Open banking APIs differ by market and by bank, fallback rates matter enormously, and the consumer experience depends on bank-side authentication flows you do not control. Refunds, recurring payments and disputes all need product answers, because the scheme-level machinery cards provide simply does not exist in the same form.
Settlement is the bright spot. Instant rails mean the money actually moves when the customer pays, which simplifies float management compared with card acquiring cycles. But reconciliation logic has to be rebuilt, because references and payer data arrive differently than card clearing files.
What to build now
Our advice to clients follows three lines. First, treat pay by bank as an additional rail with its own unit economics, not a card replacement project. Measure it against the payment methods it genuinely displaces in each market. Second, invest early in conversion analytics per bank, because the aggregate numbers hide bank-specific failures that quietly cost revenue. Third, write the dispute and refund playbook before volume arrives. The absence of chargebacks is an advantage for merchants and a customer-service challenge at the same time.
The companies getting this right are not asking whether pay by bank beats cards. They are asking which transactions in their mix are badly served today, and routing those to the better rail. That is a quieter strategy than the headlines suggest, and it is the one that makes money.