Last week we looked at Swift’s Scheme and Nexus as two different routes to the same G20 destination. Both are converging on the same first milestone of that roadmap: speed. Three-quarters of cross-border payments are settled within an hour, the G20 target says - and between instant domestic rails, Nexus’s interlinking, and Swift’s own commitments, the industry is closer to that than it’s ever been.
So it’s worth asking the obvious question before we move on to the next three G20 goals: is faster actually better?
RIX-INST in Sweden, UPI in India, PayNow in Singapore, FedNow and RTP in the US - domestic instant payment rails already move money in seconds. Nexus is built to carry that speed across borders. Swift’s Scheme commits to it across the correspondent network. Speed is no longer the hard problem in payments, it’s becoming table stakes.
That’s genuinely good news. It’s also where the story gets more complicated than “faster is better.”
An instant payment is, by design, hard to unwind. That’s the point, it removes the delay that made payments feel slow. But that same delay used to double as a safety window: time for a bank, or a customer, to notice that an account number was mistyped, that a payee didn’t match, that something about the payment looked wrong.
Take that window away without replacing it with something else, and speed doesn’t just move good payments faster. It moves misdirected payments and fraud attempts just as fast and by the time anyone notices, the money has often already cleared and moved on. Fraud losses across the EU/EEA reached €4.2 billion in 2024, and every misdirected or returned payment still costs a bank roughly €20-60 to unwind after the fact. Instant rails don’t shrink that exposure. They shrink the time available to catch it before it becomes a loss instead of a near-miss.
This is why every major speed initiative - Nexus, Swift’s Scheme, the domestic instant rails underneath both - is being built alongside a verification requirement, not instead of one. The EU’s Instant Payments Regulation is explicit about this: it mandates Verification of Payee precisely because it also mandates instant settlement. Regulators clearly don’t see speed and safety as a trade-off. They see them as a package.
“Reduce friction” gets used as a synonym for “remove checks,” and that’s the wrong reading. The friction worth removing is the kind that slows down a payment that was always going to be fine; a form asking for information the destination doesn’t even require, a manual review queue that a computer could clear in milliseconds. The friction worth keeping, and moving earlier, is the kind that catches the payment that shouldn’t go out as written.
Done well, pre-validation doesn’t feel like friction at all. It runs before the payment leaves the account: checking the payee actually matches the account it’s headed to, flagging anomalies against typical payment patterns, confirming the payment meets the requirements of wherever it’s going. For the overwhelming majority of payments, the ones that are exactly what they claim to be, it adds no perceptible delay. For the small share that aren’t, it’s the difference between a payment that gets caught and one that gets cleared and gone.
This is the layer we build. Global Payee Verification checks the payee before a payment moves - across EU VoP, UK CoP, Swift pre-validation, Kinexys/Liink Confirm, FedNow and RTP, all through one API - and anomaly detection runs alongside it, watching for the patterns a static check alone won’t catch. Neither sits in the customer’s way. Both sit ahead of the rail, whichever one the payment happens to take, so a bank isn’t choosing between “fast” and “safe” scheme by scheme.
The G20’s speed target will likely be met and the rails are already most of the way there. The more useful question for a bank isn’t “how fast can a payment arrive?” but “how fast can a payment arrive and still be one we’re confident should have been sent?” Speed without that answer isn’t progress. It’s just risk delivered faster.
Next in this series: speed is the first G20 goal. The second is arguably the one banks feel the least pressure on today, and the one customers complain about the most: knowing what a payment will cost and when it’ll actually land.