Field Notes from Brazil — Volume 2
Lucas Albuquerque Gouveia de Lima is Bytecenture's Brazil-based payment UX researcher. This is the second in a series of Field Notes from our in-market researchers across nine countries.
A scenario I run into constantly in Brazilian payment UX studies: it's a Friday night, and the user is placing three separate food delivery orders from three different restaurants to feed everyone at home. If each order requires the user to copy a Pix code, leave the delivery app, open their banking app, authenticate, paste the code, confirm, and return to the delivery app, the experience breaks three times in a row. In a high-frequency, low-ticket category, that is the failure mode that quietly kills Payment Success Rates.
The traditional Copy-and-Paste Pix flow was a brilliant solution to Brazil's instant-payments problem in 2020. It is not a brilliant solution to the repeat-purchase problem in 2026.
The Brazilian apps with the highest transaction frequency — iFood is the clearest case, and a handful of high-volume entrants are moving in the same direction — have shipped a different flow: deep integration with digital wallets like Google Pay, and the use of Payment Initiation Service Provider (PISP) protocols on the open finance rails. The mechanical change is small, but the experience change is substantial. When the customer selects Pix in one of these flows, they don't leave the app. The system either triggers biometric authentication inside the wallet or uses open finance rails to initiate the payment without breaking context. The user stays inside the buying journey from selection to confirmation. For a user placing three orders in a row, this is the difference between a frictionless habit and a tolerated chore.
There are a few things worth saying about why this matters operationally. Pix already represents a large and growing share of Brazilian e-commerce, with industry projections pointing toward roughly half of e-commerce volume in the next two years. Once a payment method passes that volume threshold, the question stops being "are we offering it?" and starts being "are we offering the fast version of it?" A platform that accepts Pix but routes users through copy-paste is, in 2026, no longer competitive on checkout. It just doesn't know it yet, because the failure pattern doesn't show up where most analytics dashboards are pointed.
The user behaviour to watch isn't fraud, and it isn't decline rates. It is silent abandonment at the moment the user leaves the app. Standard dashboards rarely surface this clearly, because the user didn't abandon the cart in the cart sense — they accepted the payment instruction, opened their banking app, and then didn't return. In several studies I've run—including a 2026 study across three delivery apps where the gap between Pix instruction and confirmation accounted for 14% of total funnel loss—the friction of switching apps significantly impacts completion rates. PISP-based and wallet-integrated flows close that gap because the user never leaves.
What is still open is which integration model wins. The Google Pay path is mature and clean but depends on a US-headquartered intermediary at a moment when Brazilian regulatory direction on data residency and competition is evolving. The native PISP path on open finance is regulator-aligned but operationally heavier for platforms to integrate at scale. Most platforms shipping into Brazil through 2026 and 2027 will probably run both, and the question of which to default to for which user segment is itself a UX decision with measurable revenue consequences.
For global commerce platforms, the practical takeaway is narrower than the broader Pix-adoption story. By mid-2026, "accepting Pix" is table stakes. The competitive surface has moved one layer up, to "accepting Pix without breaking the user out of the buying journey." That layer is where Brazilian checkout will be optimised through 2027, and where the platforms with the highest in-app frequency will set the user expectation that everyone else is then judged against.
The platforms that learn to instrument for the silent gap — the moment between Pix instruction and Pix confirmation — will be the ones making the right investment decisions over the next eighteen months. The ones still watching only fraud and decline rates will be optimising the wrong surface.