Series 2 — Prescription · Paper 1
Not a Revolution. An Alternative.
The most common mistake people make when they hear about a new payment scheme is to assume it is trying to replace the existing one. That framing is understandable — payments is a winner-takes-most market, and the history of the industry is littered with attempts to dislodge the incumbents that failed expensively and publicly.
This is not that kind of attempt. And understanding why requires being precise about what “replacing the card networks” would actually mean, versus what building a genuine alternative means — and why the second is both more achievable and, ultimately, more important.
Visa and Mastercard process tens of billions of transactions annually across virtually every country in the world. They are embedded in point-of-sale hardware, in e-commerce checkouts, in consumer habits formed over fifty years, in the financial infrastructure of every bank that issues cards and every merchant that accepts them. Displacing that is not a payments problem. It is a civilisational project. Anyone who says otherwise is not being serious.
Building a genuine alternative is a different kind of ambition. It doesn’t require winning a war. It requires building something good enough, open enough, and cheap enough that merchants and payers choose it when they have the choice — and then, steadily and deliberately, making that choice available in more contexts.
The goal is not to make card networks obsolete. It is to make them optional.
Where the choice is easiest to offer
Not every payment context is equally suited to an alternative. A tourist using a card at a foreign hotel is not the right starting point. A consumer using a stored card for a one-click online purchase is not the right starting point. These are contexts where the card network’s advantages — global acceptance, stored credentials, consumer familiarity — are at their strongest and the friction of switching is at its highest.
The right starting point is where those advantages matter least and the card network’s costs and limitations matter most. That means verticals with three characteristics: merchants with acute fee pain and high transaction volumes; payer relationships that are ongoing rather than one-off; and payment flows where the card network’s model creates friction that neither merchant nor payer actually wants.
Several verticals meet these criteria clearly. Three are worth examining in detail — not because they are the only ones, but because they illustrate the range of what becomes possible when the payment infrastructure is open rather than closed.
The bar tab
Everyone who has spent an evening at a pub or restaurant knows the tab. You settle in, you order, you order again. At some point someone asks whether you’d like to start a tab. You hand over your card. It disappears behind the bar.
That moment of handing over the card is, if you think about it, a peculiar act of trust. You have no idea what the final bill will be. The bar has no guarantee your card will clear it. The card network, designed for discrete point-of-sale transactions, has no model for what is actually happening — an ongoing relationship between two parties who have agreed, informally, to settle up later.
The solutions the industry has reached for are all workarounds. Hold the card — inconvenient and risky. Pre-authorise a fixed amount — clunky, frequently wrong. Ask for payment per round — kills the atmosphere. None of them are solutions to the actual problem. They are patches on a payment infrastructure that was never designed to carry a relationship.
The question the bar tab prompts, once you see it clearly, goes well beyond hospitality. How many commercial interactions involve an ongoing relationship — a period of time, a shared understanding, a final settlement — that the current payment infrastructure simply cannot model? The hotel stay. The car hire. The day at a spa. The open account with a trusted supplier. In every case, the payment system treats a relationship as a series of disconnected transactions, because that is all it knows how to do.
The rental deposit
Almost everyone in the UK who has rented a property has a deposit story. Money paid over at the start of a tenancy, held for months or years, and then — at the end — subject to a negotiation that can feel arbitrary, opaque, and weighted against the tenant. Deposit protection schemes exist precisely because the relationship between landlord, tenant, and deposit was so consistently fraught that regulation was required to impose a neutral party.
The deposit problem is, at its core, a payment problem. A sum of money is held pending mutual agreement that certain conditions have been met. The landlord needs confidence that the property has been returned in the agreed condition. The tenant needs confidence that their money will be returned when it has. Both need a record of what was agreed and a process for resolving disagreement that neither party controls unilaterally.
What the current system provides is a separate bureaucratic scheme, layered on top of the payment infrastructure, to do what the payment infrastructure itself should be capable of doing. The deposit is paid. It sits in a scheme. A dispute process is invoked. Weeks pass. Someone wins, someone loses, someone is dissatisfied.
The question this prompts is a natural one: what if the payment itself carried the terms? What if the conditions for release — agreed by both parties at the outset, not argued over at the end — were part of the payment authorisation rather than a separate administrative layer? The deposit would release when both parties confirmed the conditions were met. Disagreement would trigger a defined process. The neutral party would already be built into the infrastructure, not bolted on afterwards.
This is not a property-specific idea. It is an illustration of a much wider class of situation where money is held pending conditions — and where the gap between the payment and the agreement creates friction, dispute, and the need for expensive intermediaries that exist only because the payment infrastructure was never sophisticated enough to make them unnecessary.
The subscription trap
The experience is familiar enough to have become a shared cultural frustration. You sign up for something — a streaming service, a gym membership, a software subscription. At some point you decide you no longer want it. You try to cancel. The cancellation process is, by design, more difficult than the sign-up process. You navigate menus, find phone numbers, wait on hold, are offered discounts to stay. Eventually you give up, or you go to your bank and dispute the charge.
That last step — disputing the charge with the bank rather than cancelling with the merchant — is the tell. It reveals exactly where the power in the relationship sits. The merchant designed the cancellation process. The merchant controls when and how the recurring payment operates. The payer’s only real recourse is to involve a third party — the bank — and invoke a dispute mechanism that was designed for fraud, not for subscription management.
The underlying problem is that the payer never owned the recurring authorisation. They granted it at sign-up, on the merchant’s terms, and the merchant has held it ever since. The payer’s ability to modify, pause, or cancel that authorisation runs through the merchant’s systems — which are, rationally, designed to retain revenue.
What this illustrates is the difference between a payment system where the merchant owns the recurring relationship and one where the payer does. In the second model, the payer sets the terms at the outset — the amount, the frequency, the duration, the conditions under which it continues. The merchant operates within those terms. Cancellation happens through the payment scheme, not through the merchant’s customer retention process. The payer’s control is structural, not dependent on the merchant’s goodwill.
Once you see this clearly for subscriptions, it is hard not to see the same dynamic everywhere that recurring payments exist. Rent. Insurance. Utilities. In every case, the payer authorised a recurring payment on terms they didn’t set and can’t easily change. The question of who should own that authorisation — and what it would mean for the relationship if the answer were different — is one that current payment infrastructure has never been designed to ask.
What these three have in common
On the surface, a bar tab, a rental deposit, and a subscription cancellation have nothing to do with each other. They involve different merchants, different payers, different commercial relationships, and different frustrations.
What they share is more interesting than their differences. In each case, the payment infrastructure handles the money movement competently — funds transfer from one account to another — but fails to carry the relationship that surrounds it. The tab is a relationship with a duration and a ceiling. The deposit is a sum held against conditions. The subscription is an ongoing authorisation whose terms should belong to the payer. The card network treats all three as transactions, because transactions are what it was designed for.
The gap between what a payment is and what the commercial relationship around it requires is not a gap that can be closed by processing payments faster or more cheaply. It requires a payment infrastructure with a richer model of what authorisation means — one where the consent can carry context, conditions, and terms, rather than simply recording that a transaction occurred.
These three examples are not a complete picture of what that richer model makes possible. They are an invitation to think about where else that gap exists — and what would become possible if it were closed.
The journey, honestly described
The verticals above are the beachhead, not the destination. They are where the switching case is clearest, where the innovative capability is most immediately legible, and where the cold-start problem — getting enough merchants and payers on the scheme to make it useful — is most tractable.
From that beachhead, the path is one of deliberate expansion. As more merchants in more verticals join the scheme, the payer’s reason to be enrolled grows. As more payers are enrolled, the merchant’s reason to accept the scheme strengthens. Network effects in payment systems are powerful and self-reinforcing — the same dynamic that made card networks dominant makes any scheme that achieves critical mass in its target verticals progressively harder to displace.
The question of whether that trajectory eventually produces a scheme that challenges the card networks at scale is not one that needs to be answered now. It is a question that the market will answer, as it always does, based on whether the alternative is genuinely better for the people using it.
What can be said now is this: the card networks are not invulnerable. They are expensive, closed, and increasingly misaligned with the needs of the merchants and payers they serve. An alternative that is cheaper, open, and designed around the people who use it does not need to be universally adopted to be consequential. It needs to be good enough that the people who have the most to gain from switching choose to do so.
That is an achievable ambition. It is also, if the alternative is well designed and well governed, an unstoppable one.
Thomas Larsen is a cloud platform architect and engineering leader with twenty years’ experience building open infrastructure for public-sector and defence organisations. He is currently working on an open payment scheme for the UK market.