Start with the fundamentals. UPI is a domestic real-time inter-bank payment rail operated by NPCI under RBI supervision. Cards are an international payment network operated by Visa, Mastercard, and a handful of smaller schemes, with Indian banks acting as issuers under their own commercial agreements. The two rails were designed for different problems. UPI was built for low-cost ubiquity in a country with hundreds of millions of bank accounts; cards were built decades earlier for cross-border trust between strangers. Both work. They do not work the same way.
The first practical consequence is approval rate. Indian-issued Visa and Mastercard transactions to gaming-classified merchants have persistently poor approval rates, and the failure mode is at the issuing bank rather than at the acquirer. The pattern is structural: Indian banks that issue cards to consumers are generally conservative about gaming-classified MCCs, regardless of the specific operator's quality, regardless of the specific player's relationship with the bank, and regardless of the per-transaction risk rating. UPI has no equivalent failure mode. The transaction either has the funds and authenticates, or it does not — and "does not" is almost always something the player can resolve by retrying or by funding their UPI handle from a different account.
The second consequence is speed. UPI confirmation lands in sub-second windows on properly tuned integrations because the rail was designed for that. Card transactions follow a multi-step authorization-and-settlement flow that, even when nothing goes wrong, is materially slower than the wallet-to-wallet feeling Indian players are used to from the rest of their digital life. A casino cashier that takes ten seconds to confirm a card deposit while the player's UPI app already confirmed an unrelated payment in one feels broken to that player, even when the cashier is functioning correctly.
The third consequence is cost. UPI's rail-level cost structure was designed for low-cost ubiquity rather than for processor profit. Cards carry interchange, scheme fees, acquirer markup, and the various components of the card-network economic stack. For deposit volume dominated by UPI, transaction-side cost stops being a meaningful slice of operator economics. For deposit volume dominated by cards, it does not. Operators on a card-led Indian cashier who try to read their P&L line by line eventually arrive at a question: where exactly is the margin going? The honest answer is "into the international card stack and out of India," and the answer is not flattering.
The fourth consequence is mental model. Indian players think about payments in app brands rather than in rail names. They use PhonePe, Google Pay, Paytm — they do not consciously think "I am about to do a UPI transaction." A cashier that names those apps as deposit options speaks the player's language. A cashier that shows a card form asks the player to translate from their habitual mental model into a different one — for an operator they are still evaluating for trust, in a category where translation friction reads as suspicion. The Paytm and PhonePe page covers the per-app branding argument in detail.
The fifth consequence is regulatory and partner. UPI sits inside NPCI's framework, which is opinionated about how merchant onboarding, transaction reporting, and partner-side compliance should work. Cards sit inside the international scheme rules, which Indian acquirers implement with their own overlays. Operators on cards are exposed to issuing-bank-side decisions they have no visibility into; operators on UPI are exposed to NPCI-side rules that are at least public and consistently applied.
So why is anyone still leading Indian cashiers with cards? Three reasons, none of them particularly defensible. First, inertia: many international processors had cards integrated before they had UPI, and the cashier reflects what the processor has rather than what the market needs. Second, comfort: card flows are familiar to operators with international deployments who treat India as just another market. Third, the assumption that "Visa and Mastercard work everywhere" is true in a sense that does not survive contact with Indian gaming-classified merchant approval rates.
The right move for an Indian iGaming cashier in 2026 is to lead with UPI, name PhonePe and Paytm specifically, surface IMPS and Net Banking for higher-ticket flows, and configure cards as a long-tail option for the players who specifically want them. The rail mix matters because the rails matter. Cards are not the wrong rail; they are the wrong primary rail for this market. The India market page walks through the broader deployment picture.
We run UPI-led iGaming cashiers for operators serving India. Message us on Telegram if your current Indian deposit numbers do not match the market's potential.