What PayNow Gen 2 gets right, and what it risks under-valuing

When I read the plans for PayNow Gen 2, the feature that interested me most was not another way to scan a QR code or make a faster transfer. It was the proposal to attach more structured information to payments so that businesses can reconcile them automatically. That may sound like a minor back-office improvement. In my experience it is where much of the real payment problem sits.

A school can receive a tuition payment in seconds and still spend hours working out which student it belongs to. The money may come from a parent, an education agent or an overseas account held under a different name. The payer may forget to include the student number. Foreign exchange charges may leave the amount slightly short. The school may be collecting through several bank accounts at once, so the finance team logs into each portal in turn and works down the statements line by line. When a payment arrives with no reference at all, someone has to ring the family and ask. The payment rail has done its job. The finance team still has a day’s work ahead of it.

In June, the Monetary Authority of Singapore and the Association of Banks in Singapore published the first phase of a study into PayNow Gen2. They set out four areas of enhancement, one drawn from each of four themes: customer experience, business payments, network coverage and scheme-level enablers. The business payments item is the one to watch. It covers request-to-pay, expanded cross-border connectivity and structured data fields to support automated reconciliation. Feedback on the phase one findings closes on 15 August, and an implementation roadmap is due by the end of the year.

The public discussion will naturally settle on speed, reach and convenience. For businesses, context matters just as much. A transfer of SG$10,000 (US$7,815) is not especially useful if the recipient cannot tell which customer sent it, which invoice it settles or what should happen next.

The figures suggest this is not a niche concern. By the end of 2025, PayNow had around 11 million proxy registrations, covering more than nine in ten adults and some 350,000 businesses. Over the course of that year it carried about SG$154 (US$120.4 billion) billion in consumer payment value and SG$147 (US$114.9 billion)billion in business payment value. PayNow is already close to half a business rail. It is still largely judged by consumer standards.

Also Read: The next AI payments boom may happen in the back office

I spent much of my career in China, where mobile payments and immediate transfers became ordinary relatively early. That experience can make it easy to assume that once money moves instantly, the payment problem has been solved. Building a payments business across several markets has shown me the opposite. The more payment methods, bank accounts and countries a company adds, the harder it becomes to understand what is coming in. Accepting money is often the simple part. Identifying it, reconciling it and connecting it to a company’s own systems is harder. Healthcare providers must link payments to patients and treatments, software businesses to subscriptions, marketplaces to buyers, sellers and their own fees. Each new payment method can make it easier for customers to pay while making the resulting records harder to untangle.

This is also why “real time” can be a misleading description. A payment involves several different clocks. The payer receives an immediate confirmation. The recipient receives a notification. The funds become available. Settlement occurs. The company’s ledger and customer records are updated. These events do not always happen at the same time, and a business experiences all of them as one.

Immediacy is still valuable, and not only for convenience. It replaces a promise with evidence. A buyer no longer needs to send a screenshot and ask the seller to believe the payment was made, because the seller can watch the money arrive. But a company needs a further level of certainty. It has to know not only that money arrived, but who sent it, why, and which obligation it settles. A payment is not fully real time until a business’s systems can recognise it and act on it. The ideal is not merely a transfer that lands instantly. It is one that closes an invoice, updates a customer account and sends only genuine exceptions to a person.

Structured data is what makes that possible. Structured remittance information carried under the ISO 20022 standard supports reconciliation, cash forecasting and straight-through processing. A 2018 study by Payments Canada and EY estimated that inefficient payment processing cost Canadian companies between C$2.9 (US$2.08) billion and C$6.5 (US$4.66) billion a year, citing manual invoice matching, limited visibility and fragmented processes. The figures come from another market and another decade. The operational problem is familiar everywhere.

None of this is confined to the finance department. Payment friction is an economic cost. Transaction charges reduce merchants’ margins. Delayed settlement ties up working capital. Poor information forces staff to spend their time investigating payments rather than serving customers. Governments therefore have good reason to invest in national payment infrastructure, because connecting banks, setting common standards and lowering the cost of moving money reduces friction across an entire economy.

Also Read: The end of manual finance? AI agents are coming for startup payments

But a national rail is a standard, not a finished commercial product, and the division of labour is reasonably clear. Governments and financial institutions are best placed to establish standards, connect participants and maintain trust in the system. No private company could have mandated interoperability across Singapore’s banks. Commercial firms then build what sits above it: invoicing, reporting, reconciliation, accounting integrations and the particular workflows that different industries need. My own company does this work, so I have an obvious interest in that split. I would also argue it is the arrangement that has worked wherever it has been tried.

I think of a national payment system as a digital motorway. The public sector builds the road, connects the network and sets the rules. It does not need to manufacture every vehicle or run every logistics company. Better roads create more opportunities for businesses to build on top of them. A more capable PayNow should not remove the need for payment technology companies. It should give them a better foundation. As more information travels with each transfer, the infrastructure itself becomes more useful. Money no longer merely moves. It arrives with enough context for the recipient to understand and process it.

This matters more as domestic systems connect across borders. Singapore already links PayNow with Malaysia’s DuitNow, and regional cross-border QR use is rising fast. An IMF study found that such transactions grew by more than 300 per cent in Thailand and 550 per cent in Malaysia in 2024, though it notes the volumes remain small. PayNow’s own cross-border links carried around SG$371 (US$290) million in 2025, against roughly SG$301 (US$235.2) billion domestically. The direction is unmistakable and the base is tiny. Connecting national rails does not standardise invoices, customer identifiers, exchange-rate records or accounting systems. Faster regional payments may therefore increase the need for orchestration rather than reduce it. The more markets and methods a business accepts, the more it matters to see those transactions in one place and reconcile them consistently.

The first generation of instant payment systems answered a basic question: can money move cheaply and immediately? The next has to answer a harder one. Can a business understand the payment as quickly as it receives it?

I do not think the future of payments is simply that every transfer becomes instant. That will increasingly be taken for granted. The more important change is that businesses will stop treating each incoming payment as a separate administrative task. The money, its purpose and the action that follows should move together.

That is when real-time payment becomes real-time commerce.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. You can also share your perspective by submitting an article, video, podcast, or infographic.

The views expressed in this article are those of the author and do not necessarily reflect the official policy or position of e27.

Join us on WhatsApp, Instagram, Facebook, X, and LinkedIn to stay connected.