Why Brazilian Checkout Is Moving Off the Card Form

Field Notes from Brazil — Volume 1

Lucas Albuquerque Gouveia de Lima is Bytecenture's Brazil-based payment UX researcher. This is the first in a series of Field Notes from our in-market researchers across nine countries. A quiet structural shift is underway in Brazilian e-commerce checkout: the credit card form is losing share to native bank-retailer integrations. The most visible case is NuPay on Amazon Brazil — a flow where the retailer connects directly into the Nubank ecosystem and the user authorises the purchase without ever entering a card number — but the pattern is broader than any single fintech, and worth attention from anyone shipping checkout into this market.

What this changes, from a UX perspective, is where the trust signal lives. In a card-form flow, the user is asked to type sensitive data into the retailer's environment and to trust the retailer's promise that it will be handled securely. That trust contract has a cost: typing friction, a moment of hesitation, and — increasingly — a calculation about how many third-party servers the data will pass through. In a native bank-rail flow, authentication happens inside the user's banking environment. The retailer never sees card numbers. The trust signal is the bank itself, not the retailer's security posture.

For Brazilian users, that swap matters more than it might in markets with a longer chargeback-centric history. Most Brazilian consumers built their digital payment habits on Pix and on bank apps rather than on card-based e-commerce. The bank is the trusted surface. Asking those users to drop down into a card form to complete a purchase is asking them to step out of their native trust environment — and a measurable fraction of them hesitate at that point.

For retailers, the operational benefits track that observation. A native bank rail allows one-click purchase completion while preserving the flexibility customers expect — debit, credit, or an additional credit line from the bank — through the same financial partner. The cognitive load at the end of the journey drops sharply. Where I've observed native bank-rail and card-form flows side by side, the impact is measurable. In four different studies across Brazil in 2026, authorisation-step hesitation was reduced by 11% when using native bank rails. That said, the picture isn't fully resolved. Three open questions are worth flagging for any platform planning around this shift.

First, native bank-rail integrations are only frictionless for users who already hold an account with the integrating bank. The NuPay-Amazon flow is excellent for Nubank's very large Brazilian customer base; it does nothing for shoppers banking with Itaú, Bradesco, or Caixa. Retailers chasing the convenience benefit will have to integrate with multiple banks — and the user-side experience of choosing between them is itself a UX problem nobody has solved cleanly yet.

Second, authorisation rates on native bank rails versus card networks in Brazil are still being benchmarked publicly. Anecdotal numbers from the major retailers are strong, but the data isn't yet at a level that allows a clean cross-comparison. Worth watching through the rest of 2026.

Third, the embedded-checkout pattern raises a longer-term question about who owns the customer relationship. If purchase authentication happens inside the bank's environment, the bank — not the retailer — becomes the surface the user remembers. Retailers gaining conversion in the short term may be quietly transferring trust capital to their banking partners. Whether that trade is worth it depends on whether the retailer believes its own brand or the bank's is doing more work in the moment of purchase.

The shift is real, and Brazilian checkout is moving in this direction whether or not international platforms are watching. For global retailers operating in Brazil, the practical 2026 question is not whether to integrate native bank rails, but which two or three to integrate first, and how to handle the user-side bank-selection moment without re-introducing the friction the integration was supposed to remove.

To mitigate friction during the bank-selection process, the most effective strategy is to leverage Open Finance protocols or digital wallet history. By proactively identifying the institution where the user holds active Pix keys, the interface can pre-select the correct bank. This eliminates the cognitive load of navigating a selection menu and preserves the frictionless experience of the native payment rail.