Banks and financial market infrastructure
Commercial and regional banks, custodians, clearing houses and market infrastructure operators, where the integration is only finished when it can be explained to an examiner.
Continuous settlement, structured message data and availability at every hour are being asked of estates that post in batch, close for a maintenance window on Sunday, and reconcile at end of day by a process that is really a person.
Ripple Systems designs, builds and operates the integration between the XRP Ledger and the systems a payment business already runs. The ledger is the part that works on day one. Everything it has to connect to is the part that decides whether the programme reaches production.
No platform to license · No vendor fees · Built in your repositories
of in-scope cross-border traffic moved to ISO 20022 when MT coexistence ended in November 2025. The message format is settled. What the messages connect to is not.
Swift, CBPR+ cutover, 22 November 2025
Unstructured postal addresses are decommissioned for cross-border payments. Structured or hybrid becomes mandatory, and the data has to come out of systems that were never asked for it in that shape.
Swift CBPR+ roadmap
the global average cost of sending $200 across a border, against a G20 target of 3% by 2030. The cost sits in the correspondent chain, not in the message.
World Bank Remittance Prices Worldwide · confirm the current quarter before launch
The FSB's five-year review of the G20 cross-border payments roadmap finds its targets unlikely to be met, naming slow infrastructure upgrades and continued reliance on correspondent banking.
FSB, G20 Roadmap consolidated progress report, October 2025
None of them is a question about the ledger. All three are questions about what the ledger has to be joined to.
ISO 20022 is now the settled format for cross-border instructions, and the structured-address mandate lands in November 2026. Richer messages require richer data, and that data has to be produced by systems that were never asked for it in that shape.
Better visibility into a process nobody changed is what the last upgrade bought.
Transactions finalise in seconds, continuously, including at 2am on a Sunday and during the week of your core upgrade. A batch posting model with a cut-off does not. Reconciliation, break handling and reporting all inherit whichever of the two you fail to reconcile.
Running a node is straightforward. Running one with monitoring, storage planning, amendment tracking, failover and an on-call rota is an operations discipline, and it is usually discovered after go-live rather than costed before it.
A pilot is a corridor operated by people. Somebody watches it. Somebody reconciles it at the end of the day, by hand or nearly by hand. Somebody is reachable when it breaks, and the volume is low enough that they can resolve a break by looking at it. This is not a small version of production. It is a different activity that produces a similar-looking result.
Production means the core posts without an operator, reconciliation runs on a schedule, and a disagreement between ledger state and core state has a documented path rather than a person who knows what to do. It means the system is available during the maintenance window and during the core upgrade. It means an end-to-end time that has been measured at the worst hour, across every leg, including the ones a correspondent and a vendor own.
The pilot works. The production system is what does not arrive.
That distance is engineering work, and it is almost always larger than the programme costed. We build it, and we can measure it first if you would rather know before committing to a date.
Illustrative structure only. Not a client engagement, and not drawn from one.
Six of connecting a ledger to a payment estate, six of operating the result. The same twelve are the scope of an assessment, the scope of a build, and the twelve questions of the diagnostic.
Connecting a ledger to a payment estate that already has a core, a hub, cut-offs, reporting obligations and an operations team.
Running the infrastructure afterwards, on the assumption that it has to stay up, stay current, and prove what it did.
A supervisor, a CFO and an operations director are protecting different things. The engineering underneath is largely the same; the evidence each of them needs is not.
Commercial and regional banks, custodians, clearing houses and market infrastructure operators, where the integration is only finished when it can be explained to an examiner.
Payment service providers, cross-border and remittance operators, e-money institutions and fintechs, where the integration has to hold at the volumes the business plan assumes rather than the volumes the pilot ran.
Corporate treasury functions, supply chain and logistics operators, energy and utility settlement, and public registries — where the ledger is being asked to carry obligations that currently live in an ERP and a spreadsheet.
Assessment, build or operate — three ways in, of equal standing. If you already know what you have, there is no reason to be assessed before anything can be built. If you do not, four weeks of measurement is cheaper than the first wrong assumption.