Real-time payments have become one of the defining themes in the evolution of global finance. Across wholesale markets, momentum is building behind new architectures (from instant payment rails to tokenised settlement models and unified ledgers) that are promising faster, more transparent, and more efficient movement of money.

Yet, for all the progress, the reality on the ground seems to be rather uneven. Cross-border payments are still constrained by fragmented infrastructure, inconsistent standards, and legacy operating models that were never designed for a 24/7, always-on environment. As new technologies push the boundaries of what is technically possible, banks and treasuries are being forced to rethink how the manage liquidity and control risk across jurisdictions.

To help understand how these dynamics are playing out Trade Treasury Payments (TTP) spoke with Pratiksha Pathak (PP), Partner and Head of Payments at RedCompass Labs.

What is different this time

TTP: There’s a lot of momentum behind wholesale payments modernisation right now (in the realm of unified ledgers, tokenisation, central bank money, etc.). From where you sit, what feels different this time compared with previous reform cycles?

PP: This time, modernisation has moved beyond discussions and into live execution. We’re no longer just talking about standards or frameworks. We’re seeing real-world pilots and testing through initiatives like Project Agora.

The other big shift is regulatory clarity. Frameworks such as MiCA in Europe and the Genius Act in the US are providing clearer guidelines that were missing in the previous attempts at reform.

What feels different now is the combination of operational experimentation and regulatory alignment. The conditions for implementation at scale are finally starting to come together.

TTP: On the ground, though, many banks would say cross-border payments are still slow, costly, and opaque. Why is delivery proving so uneven despite the policy commitments and roadmaps?

PP: There is a clear gap between global policy ambition and on-the-ground delivery. While policy roadmaps like the G20’s set the direction, implementation still depends on fragmented infrastructures, legacy systems and uneven levels of readiness across geographies.

Interoperability is the biggest challenge for cross-border payments. Banks are still operating across multiple rails, standards, and settlement models, which limits efficiency. Progress is uneven because execution requires coordinated change across technology, regulation, and operating models, not just high-level policy commitments.

From batch to real-time operations

TTP: You’ve spoken about real-time settlement changing liquidity dynamics. What actually happens to a bank’s balance sheet when payments move from batch to continuous processing?

PP: Traditionally, bank balance sheets have been updated in batches, with liquidity positioned and adjusted around end-of-day cycles. Instant payments fundamentally change that model. The balance sheet becomes a continuously updated, real-time construct rather than a periodic snapshot.

As payments move to 24/7 rails such as RTP, FedNow, and SEPA Instant, banks lose the ability to rely on netting, deferred booking, or operational downtime to manage liquidity. Funding and liquidity decisions increasingly need to be made intraday, across currencies and systems, with far less tolerance for delay or imbalance.

This is why many banks are decoupling real-time payment decisioning from legacy core systems. Stand-in capabilities allow balances, limits, and controls to be maintained continuously, even when the core is unavailable, ensuring settlement and liquidity positions remain accurate in real time.

The overall balance sheet does not necessarily become larger, but it becomes more sensitive. Liquidity buffers must be positioned precisely, and treasury functions shift from end-of-day management to continuous liquidity orchestration.

TTP: Real-time sounds positive, but it can also expose weaknesses. What kinds of control or risk issues tend to surface when institutions switch from end-of-day processes to always-on payments?

PP: Moving to always-on, real-time payments can expose issues that were previously masked by end-of-day batching. Liquidity management becomes continuous rather than periodic, which places new demands on intraday monitoring and operational resilience.

From a control perspective, compliance and AML processes must operate in real time, with far less tolerance for manual intervention or delay. Instant settlement also reduces the ability to unwind errors or fraud once a transaction has been processed.

These challenges become even more visible on newer rails such as stablecoins or tokenised settlement models, where custody, redemption, and exception handling must operate continuously. This is why resilience mechanisms such as stand-in controls are becoming increasingly important. Banks need the ability to apply limits, checks, and authorisation logic even when legacy core systems are unavailable.

This is why resilience mechanisms such as stand-in controls are becoming increasingly important. Banks need the ability to apply limits, checks, and authorisation logic even when legacy core systems are unavailable.

Real-time payments reward robust controls and operational resilience, but they expose any weaknesses very quickly.

TTP: Governments and treasuries often struggle to keep pace with these shifts. Where do you see policy, treasury, or funding models lagging behind the technology today?

PP: What we are seeing is that technology is already capable of enabling near-instant cross-border settlement, but policy and treasury operating models are still built around legacy assumptions.

Most government and institutional treasury functions are still structured around business-day liquidity cycles, prefunding models, and end-of-day reconciliation. That mindset does not translate well into always-on settlement environments, whether through instant payment rails or tokenised settlement models. The result is that the technology can settle instantly, but the funding and governance models behind it remain slow.

Policy also continues to lag in terms of cross-border alignment. Individual jurisdictions are moving forward with frameworks but the real challenge is how these regimes interact. Stablecoin issuance, redemption, custody, and reserve treatment are still interpreted differently across markets, which creates friction for global scalability.

The gap is no longer about whether the technology works. It is about whether treasury, funding, and regulatory models can adapt quickly enough to support always-on liquidity, real-time risk controls, and global settlement certainty.

What is holding back scale

TTP: Interoperability keeps coming up as the sticking point. What does that fragmentation mean for banks trying to scale cross-border flows?

PP: Fragmentation makes scale expensive and operationally complex. Banks today are juggling multiple payment rails, message standards, and governance models, which limits the ability to deliver consistent service levels and erodes the efficiency gains that modernisation promises.

As new rails such as instant payments and blockchain-based settlement sit alongside legacy infrastructure, banks are faced with making them work together. Scaling cross-border flows will require interoperable, rail-agnostic architectures.

TTP: If you were advising a bank executive planning for 2026 and beyond, what are the first two or three foundational investments they should prioritise to make innovation actually stick?

PP: The first priority is interoperable, rail-agnostic payment infrastructure. Banks need architectures that can integrate instant payments, stablecoins, and legacy rails without constant re-engineering.

The second is operating models. Treasury, risk, compliance and operations need to be embedded from the outset, not brought in after technology decisions are made. Real-time payments require continuous change, not one-off transformation programmes.

Finally, banks should focus on targeted pilots in high-friction corridors tied to real activity. Success should be measured operationally through faster settlement, fewer reconciliation breaks and lower cost per transaction.