The sandwich solved one problem — but not the whole stack

If you've spent any time around cross-border payments over the last two years, you've heard the metaphor: the stablecoin sandwich. Fiat on either side, stablecoin in the middle. Convert, settle, convert. It's the elegant pattern that finally let businesses route around correspondent banking, T+5 settlement, and 30% pre-funding lockups.

And it worked. The sandwich made the case for stablecoins not as a crypto experiment, but as settlement infrastructure. Visa wrote about it, Stripe built around it, and a generation of fintechs raised money by explaining it. And the numbers quickly caught up with the vision. Real-world stablecoin payment volume hit roughly $400 billion in 2025, more than double the year before, and Juniper Research is now projecting that cross-border B2B stablecoin flows will reach $5 trillion by 2035.

But here's the thing about a sandwich: it's just two slices of bread and a filling. Some have tried to dress it up, offering a "double-decker" with two stablecoins or a "side" of local-currency tokens, but in the end, it's still just bread and filling. Three layers, one function — and that's not enough to run a global treasury.

The companies that move money at scale today — marketplaces paying out across 40 countries, platforms collecting in 12 currencies, enterprises rebalancing liquidity across half a dozen entities — don't need a clever bridge, they need an operating system. A bridge can get you across a river, but it can't help you navigate the city. The next $4.5 trillion in stablecoin flows won't be unlocked by a better stablecoin sandwich.

What's missing from the sandwich?

Picture a Brazilian e-commerce merchant that pays a US software vendor $50,000 for the year. The sandwich tells a clean story: BRL → USDC → USD, settled in minutes. But the reality looks different. Before any USDC exists, the merchant has to collect from its own customers, hold the funds in a properly registered local entity, and move them offshore in compliance with Brazil's FX regime. On the US side, the vendor has to receive into the right account, reconcile to the invoice, and recognize the revenue. The bridge takes two seconds. The journey around it takes weeks of setup in compartments the bridge doesn't see.

Now scale that up. Picture a SaaS company selling into 30 countries. Each market wants to pay in its own currency on its own preferred rail — Pix in Brazil, UPI in India, SEPA in Europe, ACH in the US, super-app wallets across Asia. The company has to collect locally, hold balances centrally and attributably, convert when its treasury decides to, and pay payroll, vendors, and taxes back out into a different mix of those same markets. The sandwich gives them one fast transaction. They have tens of thousands of them a month, in five layers, across thirty jurisdictions. They don't need a faster sandwich. They need a system.

When you try to run a cross-border business on stablecoin settlement alone, you soon discover there are parts of the stack it doesn't solve:

• Collection. The sandwich assumes that local collection is already solved. In practice, the hardest part of payments is collecting funds, as this involves accepting local payments at checkout, localization, dispute handling, local entities, and tax treatments.
• Named accounts. Where does the money sit before, during, and after settlement? "On a provider's balance sheet" or "in dozens of currencies with moving FX rates" are not answers enterprise CFOs can accept. Reconciliation, attribution, audit, fund segregation — none of this is solved by a bridge.
• Treasury logic. When should you convert? At what rate? What obligation is that liquidity held for? The sandwich is a single transaction — a treasury is thousands of decisions a day, and most of them aren't about the payment type or even the payment itself.
• Compliance in flow. KYC, AML, invoicing and supporting documentation, and Travel Rule checks all have to happen during execution, not in a separate system that flags issues after the money has moved.
• Local payout. Stablecoin-settled, but with local payment rails. The last mile is where everyone discovers that "instant settlement" stops being instant when a beneficiary bank in a Tier-2 corridor says "we'll process this tomorrow."

The stablecoin sandwich solves one layer. A real money-movement operation has five.

The stablecoin bento: a better model

A bento is not a sandwich with extra steps. It's a different idea entirely.

A bento has compartments — each one designed for a specific purpose, each one good on its own, but together forming a complete meal in a single box. The compartments aren't arbitrary. The rice has a place. The protein has a place. The pickle has a place. And critically: the box itself is part of the design.

That's what global money movement looks like when you build it for production:

And then the box — the part the sandwich never had: an intelligent execution layer that runs compliance, routing, and visibility across all five compartments at once. Not as a bolt-on, but as the structure that holds the bento together.

How the bento model redefines the operations conversation

The way an industry talks about its own infrastructure shapes what gets built.

When the conversation centers around a "stablecoin sandwich," every fintech optimizes for a better filling: faster bridging, tighter spreads, cheaper gas — all real improvements, but all on the same layer. This doesn't address the biggest challenges in cross-border operations.

When we talk about "stablecoin bento," the question changes. Now you're asking: Are my compartments designed together, or did I duct-tape five providers and call it a stack? You start checking whether your collection layer talks to your treasury layer, whether your compliance runs in flow or after the fact, whether your named accounts are actually yours or actually a line item on someone else's balance sheet.

The sandwich gave the industry permission to ship one good idea. The bento asks the harder question: does the whole thing fit together?

The bento test: is your stack coordinated or just assembled?

These three tests reveal whether your money movement stack is actually a bento — or just a sandwich with toppings:

1. The compartment test. Can you point to each of the five layers and name the system that owns it? If three of them come down to "we email a partner," you don't have a bento.
2. The box test. When something goes wrong in compartment 4, do compartments 1, 2, 3, and 5 know about it in real time? If not, your compliance and treasury are disconnected from execution — and that's where money gets lost.
3. The local test. Is each compartment local where it needs to be? Local collection, local accounts, local payout, local compliance context — in emerging markets, a coordinated local stack beats a generic bridge every time. That's the difference between a bento made with local ingredients and an imported frozen sandwich.

The next era of money movement

The stablecoin sandwich was the right metaphor for 2023. It made the case that programmable settlement was real, that it was faster, cheaper, more transparent. It took the conversation from "is this crypto?" to "how fast can we ship it?"

But 2026 is a different conversation. Enterprises aren't asking whether stablecoins work — they're asking whether their entire money movement operation works. Now, you can run collection, custody, conversion, treasury, payout, and compliance all as one system instead of five vendors and a spreadsheet.

This isn't a sandwich. It's a bento.

And the companies that learn how to serve the whole box — not just improve the filling — are the ones that will lead the next generation of global money movement.

Pockyt is building the stablecoin bento. Programmable money movement across local rails, named accounts, stablecoin settlement, treasury logic, and local payout, unified by one execution layer. Learn more at pockyt.io.