Following a deposit
Take a customer depositing £50 by card and follow it through.
The customer enters card details on the operator's site. The data reaches a gateway, which transmits it into the processing infrastructure. The transaction is passed to an acquirer, the financial institution holding the operator's merchant relationship. The acquirer routes the authorisation request through the relevant card scheme to the customer's issuer, the bank that gave them the card.
The issuer decides. It checks available funds, applies its own risk rules, considers the merchant category code, and returns an authorisation or a decline. That decision travels back through the same chain to the operator, which credits the customer's balance if approved.
Later, funds are settled to the operator, typically after a delay, net of fees and net of any rolling reserve the acquirer holds.
Each participant takes something, and each is a potential point of failure.
Acquirers
The acquirer holds the merchant relationship, processes transactions, bears risk if the merchant fails to meet its obligations, and is the party that decides whether to serve a gambling operator at all.
Because gambling is high risk, the set of acquirers willing to serve the sector is limited, and those that do apply the conditions described in the previous lesson. Acquirer selection therefore involves fewer options and more negotiation than in other industries, and the relationship requires active maintenance rather than being set and forgotten.
Acquirers monitor merchants continuously against chargeback and fraud thresholds set by the card schemes. Breaching those thresholds moves a merchant into remediation programmes with additional fees and reporting requirements, and persistent breach leads to termination.
Payment service providers and gateways
The distinction between these is frequently blurred in practice and worth stating.
A gateway is the technical transmission layer, handling the secure passage of transaction data between the merchant and the processing infrastructure. It is infrastructure rather than a financial relationship.
A payment service provider typically bundles gateway functionality with access to acquiring and to multiple payment methods, giving the merchant one integration and one commercial relationship instead of many. For smaller operators this is efficient. For larger ones it introduces an intermediary between the operator and the acquirer, which adds cost and reduces visibility.
Many PSPs specialise in high-risk sectors, which matters because a generalist provider may onboard a gambling operator and then exit the category later.
Orchestration
Because operators run multiple providers, something must decide which transaction goes where. That is orchestration, and it has become a standard layer in this sector.
An orchestration platform sits above the providers and routes transactions according to rules. Those rules may consider the customer's country, the currency, the card type, the issuer, the transaction value, the historical acceptance performance of each provider for similar transactions, and the cost of each route.
Its most valuable function is retry logic. When a transaction declines for a reason that suggests a different route might succeed, orchestration can retry through an alternative provider automatically. Recovering even a modest proportion of initial declines is commercially significant given what a lost first deposit costs.
It also provides failover, so that a provider outage does not stop deposits, and consolidated reporting across providers, which is otherwise a reconciliation exercise.
The build-or-buy decision here mirrors the platform decision described in the iGaming Basics course. Buying an orchestration platform is faster and brings established integrations; building gives control and avoids a fee on every transaction. And as with platforms, switching cost is the decisive consideration, because orchestration sits in the path of every payment the business processes.
Alternative payment methods
Cards are the default in many markets and are frequently not the best option in this sector.
Bank transfers, including instant payment schemes and open banking initiated payments, move money directly from the customer's account. They bypass card issuers entirely, which removes the largest single cause of gambling declines. They typically carry no chargeback rights, which removes a substantial cost and risk category. Speed varies from instant to several days depending on the scheme.
E-wallets hold a customer balance funded from other sources and allow fast payment to merchants. They are widely used in gambling because they are quick, they provide a degree of separation between the customer's bank and the gambling merchant, and they generally support fast withdrawals. Fees vary and some wallets have their own restrictions on gambling.
Local schemes dominate in specific markets and frequently outperform everything else there. These may be bank-network based, mobile-based or operated by domestic providers. Coverage of the right local method is often the difference between viable and unviable market entry.
Prepaid vouchers and cash-based methods allow customers to fund accounts without a bank relationship or without linking their bank to a gambling merchant. They are one-directional, since withdrawals cannot generally be returned to a voucher, which creates an operational requirement for an alternative payout route.
Mobile and carrier billing exists in some markets, typically with low limits and high fees, and is restricted for gambling in several jurisdictions.
Cryptocurrency is used by some operators, predominantly outside the most tightly regulated markets. It raises specific compliance questions around source of funds, sanctions screening and volatility, and most major regulated operators either avoid it or handle it under strict conditions.
The general pattern is that alternative methods frequently deliver better acceptance and lower total cost than cards in gambling, and the operators with the strongest payment performance are usually those that have invested in local method coverage rather than those that have optimised card routing hardest.
What payments actually cost
Breaking down the cost is useful because it shows where negotiation helps and where it does not.
Interchange is paid to the customer's issuer and is set by the card schemes. It is the largest component of card cost and it is not negotiable with the acquirer, since the acquirer merely passes it through. Interchange varies by card type, with commercial and premium cards costing considerably more than standard consumer debit.
Scheme fees are charged by the card networks and are similarly fixed.
Acquirer margin is the part that is genuinely negotiable, and it is a smaller share of the total than merchants often assume.
Gateway and PSP fees are typically per transaction and are negotiable.
Orchestration fees add a further per-transaction cost where an external platform is used.
Alternative method fees vary widely, from very low for some bank transfer schemes to high for vouchers and carrier billing.
Chargeback costs include the disputed amount, a fee per chargeback regardless of outcome, and the indirect cost of ratio thresholds.
Foreign exchange applies wherever the transaction currency differs from settlement currency, and the spread applied is a genuine cost that is frequently overlooked.
Reserve cost is the opportunity cost of capital held back by acquirers, which at gambling volumes is not trivial.
The practical implication is that reducing payment cost is mostly a matter of method mix and routing rather than of negotiating acquirer margin. Shifting volume from expensive card transactions to cheaper bank-based methods, and routing transactions to the cheapest provider capable of authorising them, moves the total far more than a percentage point off a margin that was never the largest component.
Reconciliation
A closing operational note, because it consumes more time than anyone expects.
Reconciliation matches three things: what the operator's platform believes happened, what the payment providers report, and what actually arrived in the bank. In a single-provider single-currency operation this is straightforward. In a multi-provider, multi-currency, multi-market operation with orchestration in the path, it is a substantial ongoing exercise.
The common sources of discrepancy are timing differences between authorisation and settlement, fees deducted at different points, currency conversion applied at rates that differ from those recorded, reserves held and released on their own schedule, chargebacks and refunds processed asynchronously, and provider reporting formats that do not align.
Operations that automate reconciliation and investigate exceptions systematically catch revenue leakage that would otherwise persist indefinitely. Operations that reconcile manually and monthly tend to discover problems long after they became expensive.
Choosing a provider mix
Given the number of participants described above, the practical question is how many relationships to hold and which.
The minimum viable position in any significant market is two live routes capable of handling the full deposit volume, so that losing one does not stop revenue. In practice most operators run more, because different providers perform better for different traffic.
The considerations when selecting are consistent. Category commitment, meaning whether the provider is genuinely established in gambling or has onboarded the sector opportunistically and may exit. Geographic coverage, since approval is market-specific. Method coverage, particularly local methods that matter in the target market. Acceptance performance, which can only be assessed with live traffic and should be tested rather than assumed from claims. Settlement terms, including timing and reserve requirements, which affect working capital directly. Technical quality, covering integration effort, API stability, reporting granularity and support responsiveness. Commercial terms, though as established these are less decisive than they appear. And financial standing, because a provider failure with funds in transit is a serious problem.
The mistake operators most often make is selecting on headline pricing. A provider with slightly better rates and materially worse acceptance costs far more than it saves, and the difference is invisible unless acceptance is measured per provider on comparable traffic.
Settlement, reserves and working capital
A dimension frequently underestimated because it does not appear in the fee schedule.
Funds do not arrive when a customer deposits. They arrive at settlement, after a delay that varies by provider and method, and net of fees and reserves. Meanwhile the operator has credited the customer's balance immediately and may have paid out winnings before the deposit funds have settled.
The result is that an operator carries a working capital position determined by settlement timing across all its providers. Growth makes this worse rather than better, because higher volume means more funds in transit at any moment.
Rolling reserves compound the effect. An acquirer holding a percentage of settlement for a defined period is holding operator capital continuously, and at gambling volumes the amount is substantial. Reserve terms are negotiable over time as a merchant establishes a track record, and reducing them is one of the more valuable outcomes of a mature provider relationship.
Currency adds a further layer, since funds settled in one currency and needed in another involve conversion cost and timing exposure.
The practical implications are that settlement terms belong in provider selection alongside pricing, that treasury needs visibility of funds in transit across providers, and that anyone modelling the cash effect of growth in this sector must account for payment float rather than assuming revenue converts to cash on the day it is earned.
Descriptors, disputes and the customer's bank statement
A small detail with disproportionate consequences.
The merchant descriptor is the text a customer sees on their bank or card statement identifying the transaction. In gambling this carries several complications at once.
If the descriptor is unrecognisable, customers dispute transactions they actually made, which generates chargebacks that cost money, count towards ratio thresholds and damage the merchant's standing with its acquirer. This is one of the most common avoidable causes of chargebacks in the sector.
If the descriptor is too explicit about gambling, some customers object on privacy grounds, particularly where statements are visible to family members. Operators have historically used obscured descriptors for this reason, and doing so directly increases dispute volume, which is the trade-off.
Some jurisdictions and some schemes now require gambling transactions to be identifiable, precisely so that customers, banks and gambling-blocking tools can recognise them. Bank-level gambling blocks, which allow a customer to instruct their bank to decline gambling transactions, depend entirely on transactions being correctly categorised. Those blocks are a genuinely useful harm reduction tool and they only work if the merchant category code and descriptor are accurate.
The defensible position is accurate categorisation with a clear, recognisable descriptor. It generates fewer disputes, supports customer-controlled blocking tools, and avoids the appearance of concealing the nature of the transaction, which regulators view unfavourably.
Where the stack is heading
Two developments worth tracking, since both change the economics described in this lesson.
Account-to-account payments, enabled by open banking frameworks and instant payment schemes, move money directly between bank accounts without card rails. For gambling this is attractive on several dimensions at once: no issuer decision to be declined by, substantially lower cost than cards, generally no chargeback rights, and immediate settlement in many schemes. Adoption varies enormously by market, and where these schemes are mature they frequently dominate.
The trade-off is that removing chargeback rights removes a consumer protection, and in a sector where disputes about whether a transaction was authorised are genuinely common, that is not a neutral change. Regulators in some markets have taken an interest.
Embedded compliance is the second, where payment infrastructure carries out functions previously handled separately: verifying identity through bank data, evidencing affordability from transaction history, and enforcing limits at the payment layer. This blurs the boundary between the payments function and the compliance function and is likely to continue doing so.
For anyone working in this area, the practical implication is that the card-centric model described through most of this lesson is not permanent, and operators whose payment strategy assumes cards remain dominant are planning for a landscape that is already shifting in several markets.
A note on integration effort
A practical consideration that shapes payment strategy more than it appears in strategy documents.
Every provider and every method requires integration work: building the connection, handling its specific data requirements, mapping its response codes, reconciling its reporting, and maintaining all of that as the provider changes its interfaces. Each additional method in each additional market adds to a permanent maintenance burden.
This is the argument for orchestration platforms and for PSPs that aggregate methods, since one integration provides access to many. It is also the reason operators sometimes offer fewer local methods than a market analysis would recommend, because the engineering capacity required to add and maintain them competes with everything else on the roadmap.
The cost is worth quantifying honestly when making method decisions. A local method delivering meaningful incremental deposit volume justifies substantial integration effort. One added because a market report mentioned it, which subsequently carries a small fraction of traffic, is a permanent maintenance obligation earning very little.
The related discipline is retiring methods that no longer perform. Payment method catalogues accumulate, and operators frequently maintain integrations for methods carrying negligible volume simply because nobody reviewed them. Periodic review of method performance against maintenance cost is unglamorous and reliably finds savings.