Subscriptions in Payments Are a Pricing Hack, Not a Model

Over the last decade, fintech has heavily borrowed from the SaaS playbook. Concepts like recurring revenue, monthly tiers, and feature unlocks at higher plans all made their way to fintech. The logic is prima facie appealing, it gives us predictable income, clean ARR metrics, and investor-friendly multiples. But the problem with this logic is that payments is not SaaS, and pricing it like SaaS creates a structural mismatch.

In SaaS, the marginal cost of serving one more customer is close to zero. The software has already been built, and adding another user costs almost nothing. Subscription pricing fits this model because the cost structure is flat. Conversely, in payments, the cost structure is variable. Every transaction has variable costs attached to it: banking partner fees, FX costs, compliance overhead and settlement costs. A subscription pricing model forces the company to absorb variable costs within a fixed price, with the hope that the customer mix averages out.

Let's see this in detail.

The cross-subsidy problem is straightforward. A customer sending $10,000 a month on a $99 subscription pays 0.99% of volume. A customer sending $100,000 a month on the same plan pays 0.10%. Here, the low-volume customer is subsidising the high-volume one. This model works when the portfolio skews toward smaller customers. But as the business matures and customers grow, the subsidy goes the wrong way. The unit economics start to deteriorate exactly when the company is successful in getting larger customers.

Revolut Business and Wise are examples of companies that have taken opposite sides of this argument. Revolut offers four subscription tiers ranging from $0 to $119 per month, and each tier has a threshold of free FX transfers. Revolut is quite successful with this model. But let's remember, Revolut is not selling a payments product. It is selling a platform with multiple products like cards, expense management, approval workflows, accounting integrations, and analytics. A significant portion of Revolut's subscribers are there primarily for the cards, spend controls and expense tooling, not for FX transfers. The average international payment usage is very low. The subscription economics work because the cost of heavy FX users is offset by subscribers who don't use the transfer feature at all.

Wise has no such cross-subsidy available. Every customer is there specifically for FX transfers. All of them will maximise usage. There is no bundle to hide behind. This is exactly why Wise charges per transfer and has not tried to offer a flat subscription. The product does not give it the customer mix that makes subscription economics viable.

The implication for a pure-play cross-border payments company is the most pointed. If you try to replicate Revolut's subscription model without Revolut's product breadth, you inherit the heavy FX users without the casual subscribers who offset them. The economics fail faster and more completely than they would at Revolut.

The moat argument is equally damaging. In SaaS, subscription pricing and product stickiness compound together. The longer a customer stays, the more data, configurations, and workflows accumulate inside the product. A customer who switches after three years has to migrate years of accumulated state. The subscription itself is a retention mechanism that gets stronger over time. In cross-border payments, this compounding does not exist. A customer who has been on a $99 monthly plan for three years is no harder to lose than one who joined three months ago. The switching cost in payments comes from integration depth, settlement dependency, and KYB re-onboarding, none of which have anything to do with the pricing model. Subscription pricing in payments retains nothing. It just packages the margin differently.

Other industries learned this the hard way. Early internet providers in the 1990s offered flat-rate unlimited access when average usage was low. When usage patterns changed, the flat-rate economics collapsed for providers that had not repriced. The gym industry offers a more instructive parallel: Planet Fitness built a business model that structurally depends on members who pay but rarely show up. The economics require a majority of inactive subscribers to subsidise the minority who actually use the facility. In payments, the high-volume customers are precisely the most valuable ones. The model that works in gyms inverts in payments: the customers you most want to retain are the ones the subscription most undercharges.

There is one scenario where subscription pricing makes structural sense in payments: when it is pricing platform capability, not transaction volume. A multi-currency treasury dashboard, an approval workflow engine, a compliance reporting tool. Here the subscription prices access to fixed-cost infrastructure, not variable-cost processing. Revolut's higher tiers are partly doing this, bundling team cards, spend controls, and API access alongside the transaction allowances. The platform pricing is defensible. The transaction pricing hidden inside the subscription is not.

Subscription pricing looks like progress. It is usually a margin problem wearing a modern costume.