Don't ask banks to change. Ask them to connect
The best financial innovations are the ones you can explain at a dinner table. Tokenized deposits pass that test. When a family member asked me to explain them, I reached for poker chips (figuratively) to illustrate the point. You hand cash to the dealer, you receive the chips, and those chips move instantly across every table in the room. Each chip represents a varying amount of cash sitting in the vault.
What changes is that you can now sit down at any table in the room and play immediately, with the chips already in your hand. You don't open a new account at each table. You don't learn each dealer's system. You don’t need to “adapt” to a new set of rules. You're just in.
Not having to learn a “new game” is what the infrastructure my team builds at Cosmos is designed for: infrastructure that fits the existing systems and helps financial institutions and governments move assets in a faster and secure way, reducing settlement time and operational cost.
When an exciting new technology emerges, the providers rush to figure out how and where they can “modernize” and “upgrade” potential customers. Having good technology is only a small part of the battle. The other part is friction, specifically, how little a participant has to change about the way they already work to start using it. A payment network grows in value with every participant that joins, which means the task is simple: make it easy for a participant to join.
Building the onboarding path for the Cosmos Tokenization Suite (CTS) has taught me that “adopting new technology” is not a simple request for large financial institutions. Every time you ask a participant to take on new systems, you set off a vendor audit, pull in their core engineering team, and ask a senior executive to put their reputation on the line. That gets you a string of isolated pilots, not a working network. Asking a bank, especially those banks still running on mainframe cores, to replace parts of that core just to join is a multi-year conversation, not a one-quarter one.
McKinsey recently outlined a great example of this: interoperability is one of the primary constraints on tokenized-deposit growth, and not because the technology is immature. The hard part is coordination. Before tokens from different banks can work together, the banks have to agree on a shared rulebook: who can redeem a token and when, how compliance and settlement finality are handled, how disputes get resolved. The institutions writing those rules together are also competing with each other. McKinsey's own read is that interoperability will be decided less by the technical protocol and more by whose rulebook collects enough signatures.
Beyond the coordination of rules, there is a second hurdle that compounds the first. A payment network loses its momentum the moment membership requires a technology overhaul. The industry often tries to solve fragmentation by demanding everyone migrate to a single platform, which is a strategy that will never scale.
So, we at Cosmos built the opposite of a rip-and-replace, and we expect payment networks to follow. CTS meets each participant where they already are: a modern bank may hand us a REST API; a legacy bank may simply send us a nightly file of transactions in whatever format they already use. The instinct may be to build for the most capable bank in the room. But the reality is the opposite. A network's reach is defined by who can join it, not who leads it. So you build for the participant with the most constraints because their limitations define how wide the network can grow
In CTS, each enrolled account gets a live on-chain twin kept in sync with the bank's existing core, plugged into the settlement rails and compliance checks it already runs. A bank joins on what it already has. It connects once, then settles with any counterparty on any network as they come online. The cost falls out of that naturally: no migration to fund, no per-counterparty integration to staff, fewer intermediaries taking a slice of the pie.
That's why "you'll be faster" was never the pitch that moves an institution. The one that does is "you can join without changing what you run, and reach the whole system once you have". Ask a participant to adopt new technology to get in, and you've rebuilt the very barrier you were supposed to remove.
None of this is new. Visa grew because banks could license into something that already worked instead of each building their own. The networks that win are the ones existing participants can join without re-tooling. That's the bet at Cosmos: interoperable infrastructure that runs on what an institution already has. Onboard people on their own terms, and the network grows. Make them adopt something new first, and it won't. Choose carefully!
