What seven years of building cross-border payments has taught us

Cross-border payments succeed when money reaches an ordinary account or wallet that the recipient already uses, at a cost and a speed that match the domestic experience around it. Everything else is detail.

We tell customer stories often, because they are true and because people remember them. But a story is a single data point dressed up as an argument. Tell enough of them and a reader walks away with a warm impression and no transferable knowledge, which is not useful to a partner deciding where to invest, or to a fintech founder working out why one market opened easily and another did not.

So this post does something different. Instead of narrating what happened at Silent Roar Media, Rocket Remit, Bless Payments or bKash, we set the stories aside and ask what they have in common. Six patterns emerged, and each traces back to the theme we keep returning to in this series: interoperability and inclusion. A payment succeeds not because it moves fast in isolation, but because it lands inside a system the recipient already trusts.

Different customers, different corridors — the same six lessons, every time.

Pattern one: cross-border payments are becoming business infrastructure, not just a finance function

For most of the industry’s history, cross-border payments meant remittances: a person sending money home to family. That framing is now too narrow. Silent Roar Media, a platform that pays creators across borders, reduced its minimum payout threshold from 250 dollars to 50 dollars, cut its payout fees by 80%, and processed more than one million dollars in payout volume within a year, reaching creators in more than seven markets.

None of those numbers describe a remittance corridor. They describe a business decision: a platform lowering the threshold at which it becomes economical to pay someone, because the payment rail no longer imposes a fixed cost that only makes sense at scale. Cross-border payment infrastructure stops being a treasury afterthought and starts shaping product decisions: who a platform can pay, how often, and at what minimum size. For any founder building a marketplace, a creator platform or a gig-work app with international reach, this is the pattern to plan around first, because it changes unit economics before it changes anything else.

Pattern two: wallets have become the preferred destination

A remittance provider serving South Asian corridors, now delivers 98.7% of its payouts into mobile wallets. When it expanded wallet connectivity, transaction growth followed at 439%.

Those two figures sit next to each other for a reason. Growth did not come from adding more countries or more marketing spend. It came from meeting recipients where their money already lives. In many of the markets we operate in, a mobile wallet is not a secondary option next to a bank account; for a large share of the population it is the primary account. Building payout rails that assume a bank account as the default, with wallets as an exception to handle later, gets the market backwards. The 439% figure is what happens when a provider corrects that assumption instead.

Pattern three: scale comes from interoperability, not corridor-by-corridor expansion

Bless Payments expanded into more than 20 countries and now reaches more than 2,000 banks and wallets across 22 countries, with payments delivered in under 30 seconds.

The traditional model for international expansion is additive: sign a partner in country A, build the integration, repeat for country B, and keep repeating until the map fills in. That model does not produce 2,000 endpoints in a reasonable time. It produces a handful of well-serviced corridors and a long tail of markets that never get reached, because each new corridor carries its own negotiation, its own technical integration and its own compliance review. Reaching thousands of endpoints requires the opposite approach: build the interoperability layer once, then plug new banks and wallets into a system that already knows how to route, convert and reconcile. The under-30-second delivery time is not a separate achievement from the endpoint count; it is a consequence of it. A network built corridor by corridor tends to slow down as it grows. A network built on a common interoperability layer tends to hold its speed, because the underlying mechanics do not change with each new endpoint.

Pattern four: domestic payment experiences have become the benchmark for cross-border

bKash, Bangladesh’s mobile financial service, enables remittances from more than 60 countries, serves more than 80 million users, processes more than 1,000 transactions per second, and delivers 99.9% of payments in under 30 seconds.

Those figures read like domestic payment infrastructure statistics, not remittance statistics, and that is exactly the point. Recipients in Bangladesh do not experience an incoming international transfer as a special category of payment that arrives on a different timeline from the money already in their bKash account. It arrives at the same speed, through the same interface, with the same reliability. This is the standard the whole industry is now judged against, whether or not it explicitly says so. Senders used to accept that the international leg of a transfer would be slower and less predictable than a domestic one. That tolerance is disappearing, because platforms like bKash have shown that 99.9% of cross-border payments can arrive in under 30 seconds when the domestic leg is engineered to the same standard as the rest of the system. Anyone building a cross-border product now has to design against domestic expectations, not against the historical baseline of international transfers.

Pattern five: local connectivity matters as much as global reach

A leading remittance player focused their expansion on corridors that together represent 44% of the country’s outbound remittances, and did so using TerraPay’s local banking, wallet and compliance capabilities in the destination markets.

It is tempting to read global reach as the headline metric: how many countries, how many currencies, how many endpoints. But reach without local depth produces a payment that arrives in the right country and stops there, sitting in an account the recipient cannot easily access, or converted at a rate that erodes the value of the transfer. Their choice to concentrate on the corridors carrying 44% of outbound volume, and to lean on local banking, wallet and compliance infrastructure in those destinations, reflects a more disciplined view of expansion: reach matters only as far as the local connection at the far end can absorb it. A payment that clears international rails but fails a local compliance check on arrival has not actually reached anyone. This is a corrective to the instinct, common among founders scaling internationally, to chase country count over depth in the corridors that carry the most volume.

Pattern six: cross-border infrastructure is powering new digital economies

The pattern running across creator payouts, marketplace settlements and agricultural export payments is that cross-border money movement no longer serves one population. It serves creators being paid by platforms headquartered abroad, sellers on marketplaces with buyers in other countries, exporters settling with overseas buyers, and communities that were previously excluded from formal payment rails altogether.

This matters because it changes who the industry is building for. A rail designed only for migrant remittances optimises for a single, well-understood flow: an individual sending money to family, on a predictable schedule, in modest amounts. A rail designed for creator payouts, marketplace settlements and export trade has to handle variable amounts, irregular timing and business-grade reporting, often with many more counterparties per sender. The providers succeeding in this environment built infrastructure flexible enough to serve both cases without maintaining two separate systems, a pattern we explore at greater length in the endpoint economy, where the shift from serving people to serving endpoints becomes explicit.

The through-line: interoperability

Read the six patterns together and one idea sits underneath all of them. Cross-border payments succeed when they stop treating the border as the defining feature of the transaction and start treating the destination account, whatever form it takes, as the defining feature instead. A wallet, a bank account and a business settlement account are all just endpoints. The providers that grew fastest in the evidence above built once for interoperability across those endpoints, rather than building repeatedly for each new corridor.

This is not an argument for TerraPay’s technology in isolation. It is an argument about how the industry should evaluate any cross-border payment provider: not by the number of countries listed on a homepage, but by how many of those countries’ domestic payment systems the provider can actually reach, at domestic speed, at a cost that makes small and frequent payments viable. We asked a group of bank leaders how they see this shift from the correspondent banking side, and their answers, gathered in what bank leaders told us, confirm the same direction of travel from the other end of the pipe.

Seven years in, the lesson is not that cross-border payments have gotten faster, though they have. It is that the definition of success has changed, from moving money across a border to removing the border from the recipient’s experience altogether.