Philip Belamant: Building the Infrastructure Behind Modern Consumer Credit

Executive Summary

Zilch is usually filed under buy-now-pay-later. Philip Belamant's own idea, as recorded in profiles of the company, was different: a payments network in which the merchant and the advertiser, not the customer, fund the cost of credit. This article argues that the distinctive feature of the model is where it places the credit decision, which is at the moment of payment, and who it asks to pay for it. That placement is why consumer credit can be described as increasingly infrastructure: a layer built into the payment experience, with its own licences and, since July 2026, its own regulatory regime in the United Kingdom.

Zilch's financials are cited as reported, for the year to 31 March 2025, with their limits stated.

Introduction: Credit Is Not Only a Product

The conventional framing of Zilch is a fintech lender that lets people split purchases. That framing is directionally true and analytically thin. It treats credit as a product the customer chooses, priced to the customer, and judged by the customer's willingness to borrow.

The more useful frame follows the money. In Belamant's design, the customer's repayment is the smaller part of the economics, and the larger part is paid by the merchant whose product is being bought and by advertisers. That changes the incentives of the credit relationship, and it changes the nature of the system. A lender earns on the borrower; a payment layer earns on the flow it carries.

This article follows the structure of The Other Markets: a system, a founder whose path crosses markets and a mechanism examined against evidence. It does not repeat the company's detailed infrastructure, the wider argument that consumer credit is a technology-driven systems problem, or the company's operating playbook. Those belong to later pieces in this cluster. It also leaves aside extended personal biography.

Who Is Philip Belamant, and Why Does the Three-Market Lens Matter?

Zilch is based in the United Kingdom, its founder is South African, and it launched in the United States in 2022. Those three facts are the three-market lens of this piece. In 2022, IT News Africa described him as "the South African serial entrepreneur" and said that after developing financial technologies in South Africa, he relocated to London, where he co-founded Zilch.

Zilch was founded in London in August 2018, and Fintech Magazine's profile names Philip Belamant as co-founder and chief executive. IT News Africa also describes him as a co-founder. This article does not credit the model to him alone.

The earlier career matters to the argument for one reason. Fintech Magazine's profile says he established PBel in Johannesburg in February 2006, and IT News Africa describes it as a branchless banking service for consumers who had no bank accounts and kept their funds in cash to pay for bills and essentials. A founder who first built payments for people outside the banking system, then built consumer credit inside a regulated British market, has seen two very different settings for the same problem: how to let people pay when the conventional instrument does not fit.

That is the sense in which the diaspora operator thesis applies here. The founders in this cluster are fluent in more than one system, and that fluency shows up in what they treat as a design choice rather than a given.

What Problem Did Belamant Identify?

The problem is friction and cost in consumer credit, and specifically who bears it. When Zilch launched in the United States, its 17 May 2022 release positioned the product against the fees and late charges consumers pay on conventional credit, quoting Belamant on the point: in 2020, US consumers paid $120 billion in fees and late charges to credit cards. The release's title offered instalment plans with no interest or late charges and with cashback rewards.

The cost of credit does not disappear in such a model. It moves. Fintech Magazine's profile records the founding idea as an "ad-subsidised payments network," in which the customer repays interest-free through a virtual Mastercard, and the profile records the idea as the "Google-isation of payments." The analogy is to a search engine, which charges the user nothing and earns from advertisers. In Zilch's case the parallel payers are merchants and advertisers.

The friction Belamant addressed is therefore two-sided. For customers, it is the cost and penalty structure of conventional credit. For merchants, it is the cost of acquiring and converting customers. A model that ties the second to the first can, in principle, reduce the cost the customer sees by routing it through the merchant's marketing budget.

The problem he identified is a mismatch of who pays and who benefits, and the design answer is a payment layer that reassigns the cost.

What Did Zilch Change?

The change is in the architecture of the transaction. In a conventional credit relationship, the customer applies, is approved and then spends. In Zilch's model as described, the credit sits inside the payment: the customer pays through a card that Zilch provides, and the repayment schedule attaches to the purchase. The decision about credit is made at the payment, and the merchant's payment of a fee turns the transaction into one in which both sides have a stake.

The revenue structure follows. Fintech Magazine's reporting identifies the company's income as performance-based affiliate commissions from merchants and advertising fees, rather than interest on customer balances. BusinessCloud reported the figures for the year to 31 March 2025, which are summarised below. They are cited as reported, since this article did not review the filed accounts.

The model's logic is easiest to see by comparison with the corridor logic examined in the piece on diaspora payment networks. A remittance network earns on money that moves between people in different places. Zilch earns on money that moves between a customer and a merchant at the point of sale, and the payer of the fee is the merchant. Both are payment layers, but they place the economics in different parts of the flow.

A payment layer that carries credit changes who has to be persuaded: the merchant becomes a customer of the layer, and the shopper becomes the means by which the merchant's spending is converted.

Why Is Consumer Credit Becoming Infrastructure?

The analytical formula, from what an idea says to its economic consequence, is useful here. What it says: credit can be offered at the point of payment rather than before it. What it means: the credit decision becomes a function of the payment layer, not a separate relationship. Who benefits: merchants that convert more sales, and customers that avoid the cost structure of conventional credit. The economic consequence is that credit capacity becomes something a payment system provides, in the way it provides settlement or fraud checks.

Two features give the claim weight. The first is licensing. Zilch secured a consumer credit authorisation from the Financial Conduct Authority, as The Paypers reported in November 2020, and announced a Financial Conduct Authority payments licence on 10 December 2025. The two licences cover different activities, and between them they cover both the credit and the payment sides of the model. A payment layer that carries credit has to be authorised for both.

The second is regulation of the category itself. The FCA began regulating Deferred Payment Credit on 15 July 2026, under rules set out in policy statement PS26/1, published on 11 February 2026. The policy statement puts deferred payment credit lending at £0.06 billion in 2017 and over £13 billion in 2024. A product that grows to that scale and is brought inside a regulatory perimeter is moving from a feature to a layer of the financial system.

The comparison with other regulated payment layers in this cluster is instructive. The Flutterwave profile examines a payments layer in African markets, and the Esusu profile examines a credit layer built from rent data. Zilch sits in a third position: credit decided at the point of payment, in a regulated consumer market.

Consumer credit is becoming infrastructure to the extent that it is provided by, and regulated as, part of the payment itself.

What the Reported Numbers Show, and What They Cannot

Revenue, net loss and gross merchandise value below are for the year to 31 March 2025, and were reported by the press from the company's filed accounts. The provisions figure is cited as City AM reported it. All are cited as reported. They describe a year that ended before the regulatory change above and before the 2026 events that attracted the most attention, and they should not be read as a statement of the company's current position.

Sources for the table: revenue, net loss and gross merchandise value come from BusinessCloud and Fintech Magazine (15 August 2025), which also gives the 79 percent loss reduction and the 73 percent growth in gross merchandise value. The provisions figure comes from City AM (13 August 2025). The two prior-year figures in the table are calculations from the reported percentages, not reported numbers.

The figures support three observations. Revenue nearly doubled in a year, which fits a model that earns on volume. The net loss narrowed sharply, which fits a model approaching scale economics. And the credit-loss provision shows that credit risk remains on the company's own balance sheet, even where the customer pays no interest.

They do not show that the model is profitable, that the loss reduction will continue or that credit quality will hold through a downturn. They also do not allow a comparison with competitors, because the sources reviewed do not report like-for-like figures. Where user numbers appear in the press, they differ by date and by definition, so this article does not state a current user count.

The numbers describe a model growing quickly with a narrowing loss, and they leave the question of durability open.

Where the Infrastructure Claim Stops

The claim has a boundary. A payment layer that carries credit still carries credit risk. The provision above is the visible form of that risk. If merchants reduce what they pay in affiliate commissions or advertising, the subsidy that allows interest-free repayment weakens, and the model would need to rely more on other income. The sources reviewed do not allow a judgement on how sensitive the model is to that change.

A second boundary is regulatory. The new regime brings the category inside a perimeter, which can strengthen the case that it is infrastructure. It also means that rules, not only market forces, now shape the product. This article reports the regime's existence and date and does not assess the detail of its requirements.

A third boundary is evidence. The financial figures are taken from press reporting of the company's filed accounts, which this publication did not review directly, so they are labelled as reported. The licences and the regulatory regime are the stronger evidence for the infrastructure reading, but they are documented unevenly: the regime by the regulator's own policy statement, the payments licence by Zilch's own release and the consumer credit authorisation by 2020 trade reporting. This publication did not check the FCA register.

The reading holds as an analytical frame, and its limits are the same limits that apply to any payment layer: the economics depend on who is willing to pay, and the licence depends on continuing to meet the regulator's standards.

Why This Matters for Fintech Operators and Capital Allocators

For operators building consumer finance products, the relevant lesson is about placement. The credit decision can sit before the payment, as a separate product, or inside it, as a feature of the payment layer. Placing it inside changes who the buyer is, because the merchant becomes a paying customer. Operators should ask which party's budget the model draws on and how reliable that budget is.

For merchants, the question is whether paying a commission to a credit-carrying payment layer delivers more value than the alternatives, including its own financing options. For regulators and policy teams, the case demonstrates how a product that grew quickly outside a perimeter was then brought inside one. For investors and allocators, the figures reported for the year to March 2025 show growth and a narrowing loss, and the provision shows risk retained. The key diligence questions are the durability of merchant-funded economics and credit performance through a full cycle.

The cluster's wider thesis is that founders fluent in more than one system build layers where others see products. Belamant's path, from payments for people outside the banking system to consumer credit in a regulated British market, is one illustration, and the evidence in this article is the licence stack and the reported accounts. The LemFi profile shows the same pattern from the remittance side.

Key Takeaways

• Zilch's distinctive feature is where it places the credit decision: at the payment, funded by merchants and advertisers rather than by interest on the customer.
• The three-market lens explains the model: a South African founder, a UK base and a United States launch in 2022.
• Licences and regulation support the infrastructure reading: consumer credit authorisation, a payments licence announced in December 2025 and the FCA's Deferred Payment Credit regime from 15 July 2026.
• The reported figures describe growth with retained risk: revenue of £110.3 million and a net loss of £10.5 million for the year to 31 March 2025, and £27.4 million of credit-loss provisions, as reported.
• The durability question is who keeps paying: the model depends on merchant and advertiser spending, and the sources reviewed do not measure that sensitivity.

Conclusion

Philip Belamant's contribution is an attempt to make consumer credit a programmable layer of the payment experience, in which the credit decision, the licence and the subsidy all sit inside the payment. The evidence for that reading is the licences Zilch holds, the regime now governing its category and the accounts reported for the year to March 2025. The evidence against overreading it is that credit risk remains on the balance sheet and the model's funding depends on merchants and advertisers continuing to pay.

What the case teaches is that infrastructure is defined by where a function sits and who funds it. Credit placed at the point of payment, with the merchant as payer, is a different object from credit offered to a customer before they shop, and the difference is the subject of this article.