The True Cost of Acceptance: Interchange, Scheme Fees and the Margin You Don’t See
Ask most finance leaders what they pay to accept card payments and you will get a single number. A percentage, sometimes with a fixed component, quoted as the merchant service fee. It is the figure that gets negotiated at renewal, compared against a competitor's quote, and reported upward as "our payments cost".
It is also the least useful number in the whole arrangement.
That single rate is an average sitting on top of a stack of distinct costs, each set by a different party, each moving independently, and each responding to different levers. Some of those costs are effectively fixed by regulation. Some are set by the card networks and change twice a year without negotiation. Some are pure margin for your provider. And a surprisingly large share is determined not by any rate at all, but by the mix of cards your customers happen to use and the routing decisions made, or not made, on each transaction.
The finance teams that materially reduce their cost of payment acceptance are almost never the ones that negotiated hardest on the headline rate. They are the ones that understood the stack line by line, and then worked on the parts of it that actually move.
This article takes that stack apart. It is written for a finance audience, CFOs, commercial finance, ecommerce and procurement, and the goal is to leave you able to read your own statement and know which costs are negotiable, which are structural, and where the real savings sit. It is worth adding that in Australia the ground is shifting under all of this right now, and the reforms landing in October 2026 change which of these levers matter most.
The three-layer stack hiding inside one rate
Whatever pricing model you are on, the cost of accepting a card payment is built from three layers.
The first layer is interchange. This is a fee the merchant's acquirer pays to the bank that issued the customer's card, on every transaction. Neither you nor your acquirer sets it; the card networks publish interchange schedules, and it is the single largest component of acceptance cost for most merchants. Interchange exists to share the cost of the card-issuing business between the issuer and the acquirer, on the logic that both sides benefit from the payment. Crucially, it varies enormously by card type: a basic domestic debit card carries a very different interchange rate from a premium consumer credit card loaded with rewards, which is different again from a commercial card or a foreign-issued card.
The second layer is scheme fees. These are the fees the card networks, Visa, Mastercard, eftpos, charge for use of their rails. They are numerous, individually small, and notoriously opaque: assessment fees, authorisation fees, cross-border fees, data and settlement fees, and a long tail of behavioural fees that penalise things like low approval rates or high refund ratios. Scheme fees are not negotiable by the merchant, they change on the networks' own schedule, and they are the layer even sophisticated finance teams understand least well, because the billing detail is genuinely hard to parse.
The third layer is acquirer margin. This is what your provider keeps for processing, settling and servicing the transaction. It is the only layer that is fully negotiable, and it is usually the smallest of the three. This is the central irony of acceptance cost: the part everyone negotiates is the part with the least room in it, while the two larger layers get treated as fixed background noise.
Understanding this structure is what turns a merchant service fee from a single number you accept into a set of costs you can act on.
Why the pricing model you are on changes what you can see
How these three layers are presented to you depends entirely on your pricing model, and the model is often chosen for you by default.
Under blended pricing, the acquirer collapses all three layers, across all card types, into a single rate. Every transaction costs the same headline percentage regardless of whether the customer used a cheap domestic debit card or an expensive foreign premium credit card. Blended pricing is simple and predictable, which is why smaller merchants are often placed on it. Its cost is total opacity: you cannot see interchange, you cannot see scheme fees, and you cannot see the margin, so you have no way of knowing whether the blend is fair or where your cost is actually coming from. You also carry the averaging risk, if your card mix shifts toward expensive cards, your cost rises, but if it shifts toward cheap ones, the acquirer keeps the difference.
Under interchange-plus pricing, the statement separates the layers: interchange is passed through at cost, scheme fees are passed through, and the acquirer's margin is shown explicitly as the "plus". This is more complex to read but radically more transparent. You can see exactly what interchange you paid, which means you can see your card mix; you can see scheme fees, which means you can scrutinise them; and you can see the margin, which is the only number you should be negotiating.
For any merchant of meaningful volume, moving from blended to interchange-plus is frequently the single most valuable step in payments cost optimisation, not because it lowers any individual fee, but because it makes the costs visible enough to manage. You cannot optimise what you cannot see, and blended pricing is designed, intentionally or not, so that you cannot see.
The Australian picture is changing fast, and mostly in merchants' favour.
This is the drafting-sensitive part, because Australia's card payment rules are mid-reform, so it is worth being precise about where things actually stand as of mid-2026.
On 31 March 2026 the Reserve Bank of Australia published the Conclusions Paper for its Review of Merchant Card Payment Costs and Surcharging, the final phase of a review that began with an Issues Paper in October 2024. The decisions are substantial and they bear directly on every layer of the stack.
Interchange fees in Australia are set within caps the RBA regulates, and those caps are being reset. The RBA is lowering the caps on domestic-issued card transactions, particularly on consumer credit cards, retaining the existing cap on commercial credit cards, and introducing a cap on foreign-issued card transactions for the first time. The RBA estimates these changes will lower merchants' wholesale card payment costs by around $910 million per year. Small merchants are expected to benefit most, because they tend to pay fees at or near the current caps, while large merchants benefit particularly from the new cap on expensive foreign-issued cards.
On scheme fees, rather than capping them the RBA has set expectations for the networks to simplify their fee schedules, publish their interchange and scheme fees, and justify increases, explicitly to make scheme fees easier for providers to understand, reconcile and scrutinise. For a finance team, that is a signal: the layer that has been impossible to interrogate is being pushed toward daylight.
On transparency more broadly, acquirers will be required to publish their merchant service fees and a measure of interchange pass-through, so that merchants can see which providers actually pass on the interchange savings rather than pocketing them. The RBA has been explicit that it will monitor pass-through and highlight acquirers that fail to pass the reductions on.
Most of these changes take effect on 1 October 2026, with the interchange cap on foreign cards and some transparency measures following on 1 April 2027.
Surcharging: the lever that is about to disappear
The most visible change is to surcharging, and it reframes the entire cost conversation.
From 1 October 2026 the RBA will lift its prohibition on "no-surcharge" rules across the designated eftpos, Mastercard and Visa networks, for debit, prepaid and credit cards. In plain terms: the networks are expected to reimpose no-surcharge rules shortly afterwards, so the practical effect is that merchants will no longer be able to pass card acceptance costs to customers as a surcharge. The RBA has said that if surcharging persists after the prohibition is lifted, it could recommend the Government legislate a ban.
The RBA's reasoning is worth understanding, because it explains why this is happening. It found that the surcharging framework, now more than two decades old, had stopped doing its job of steering consumers toward cheaper payment methods, surcharging had roughly doubled since 2022, single-rate plans meant many merchants surcharged all cards identically, and the rules had become hard to enforce. It estimates consumers pay around $1.6 billion of a total $1.8 billion in card surcharges each year on the designated networks.
The consequence for finance teams is direct and, for many, uncomfortable. If your organisation currently surcharges, acceptance cost has been a pass-through, an inconvenience to administer, but not a cost you carried. After October 2026 it becomes an absorbed cost, sitting in your margin alongside every other cost of doing business. The RBA itself anticipates that the roughly 16 per cent of merchants that currently surcharge may lift their advertised prices to cover it.
That single change is what converts acceptance cost from a back-office administrative line into a margin line that belongs on the CFO's desk. A fee you recover at checkout is worth understanding. A fee you absorb into margin is worth managing precisely.
The savings are in the mix and the routing, not the rate
Here is the part that most surprises finance teams: once acceptance cost is understood as a stack, the largest savings usually come not from renegotiating any rate, but from changing which fees apply in the first place.
Two levers dominate, and together they are really about payment mix, the composition of how your customers pay.
The first is card mix. Because interchange varies so much by card type, your effective cost is heavily determined by what your customers pay with, and by how your checkout is designed. Steering, where still permitted, and the design of payment options can shift mix. So can the simple discipline of understanding it: a business that discovers a large share of its cost comes from foreign-issued or premium credit cards can make deliberate commercial decisions about those flows, which is impossible under blended pricing where the mix is invisible.
The second, and for many Australian merchants the larger one, is least-cost routing. Around 90 per cent of debit cards issued in Australia are dual-network debit cards, which carry both an eftpos and an international scheme network and can be processed down either. Least-cost routing sends the transaction down the cheaper network. The RBA's own analysis found that the cost of accepting debit transactions was nearly 20 per cent lower for merchants with least-cost routing turned on than for those with it off.
On least-cost routing, the RBA's current position is deliberately measured, and this is where writing from old assumptions would mislead. In the March 2026 Conclusions Paper the Board decided to maintain the existing expectations-based approach for in-person, card-present transactions, acquirers are expected to offer and promote it, and large acquirers report take-up to the RBA every six months, rather than mandating it. As of December 2025, 84 per cent of merchants had it enabled for in-person transactions. Least-cost routing for online and mobile-wallet transactions was explicitly not mandated; it was deferred to the further consultation expected in mid-2026. So for a merchant today, in-person routing is a lever largely in your own hands, and online routing is an area still in flux, both worth acting on, neither safe to assume is switched on by default.
The point is that mix and routing are operational decisions, not commercial negotiations. They do not require your acquirer's agreement. And they typically move cost by more than the margin you could win back in a hard renegotiation.
The costs that hide at the edges: cross-border and currency
Two further costs deserve specific attention because they are both large and frequently misattributed.
Cross-border fees apply when the card is issued in a different country from the merchant, and they stack: higher interchange (until the new foreign-card cap applies), additional cross-border scheme fees, and often a wider acquirer margin. For any business with international customers, ecommerce, travel, education, professional services, foreign cards can be a materially more expensive slice of the book, and it is often invisible under blended pricing. The RBA's new cap on foreign-issued card interchange from April 2027 will help, but it does not touch the scheme-fee and margin components.
Dynamic currency conversion is the related trap. When a foreign cardholder is offered the choice to pay in their home currency at the point of sale, the conversion is done at a marked-up rate, and that margin is shared between parties in the chain. It is frequently presented as a revenue opportunity for the merchant, and sometimes it genuinely is, but it also adds cost and complexity to reconciliation and can worsen the customer experience. It should be a deliberate, measured decision, not a default switched on because a provider suggested it.
Both of these illustrate the same principle. Cost per transaction is not one number; it is a distribution, and the expensive tail of that distribution, foreign cards, premium cards, converted transactions, is where disproportionate cost hides. A blended rate averages that tail into invisibility. Understanding the distribution is what lets you act on it.
What Australian finance leaders should be asking now
The reforms make the next twelve months an unusually good time to do this work, because the levers are moving and the transparency that makes them legible is arriving at the same time. A short set of questions cuts to it.
Are we on blended or interchange-plus pricing, and if blended, what would our statement look like unbundled? If you cannot see interchange separately, you cannot see your card mix, and you are managing blind.
What is our effective cost per transaction by card type, not on average? The average conceals the expensive tail, and the expensive tail is where the savings are.
Is least-cost routing enabled for our in-person transactions, and what is our position on online routing ahead of the RBA's mid-2026 consultation? Given the RBA's own ~20 per cent finding, this is rarely not worth doing.
When surcharging ends in October 2026, what does absorbing our current acceptance cost do to margin, by product and channel? For anyone currently surcharging, this is a budgeting question that needs answering before October, not after.
And when interchange caps fall, how will we verify our acquirer has passed the reduction through in full, rather than widening its own margin? The RBA is publishing pass-through data precisely so merchants can check, but someone has to actually check.
Getting acceptance cost under control
The outcome worth aiming for is not a lower headline rate. It is a finance function that can read its own acceptance cost the way it reads any other cost base: by component, by driver, and by where the money actually goes. From there the actions follow naturally, unbundle the pricing so the stack is visible, understand the card mix rather than averaging it away, switch on the routing levers that are already in your control, treat cross-border and conversion as deliberate decisions, and hold your acquirer to the pass-through the reforms require.
Getting there is more analytical than commercial. It depends on reading settlement and fee data at transaction level, mapping cost to card type and channel, and modelling the October 2026 changes against your actual book rather than against a market average. That is work most finance teams are not resourced to do mid-close, which is precisely why it tends not to happen until a reform forces it.
This is where specialist help earns its place. Conexxia works with Australian finance and ecommerce teams on payments cost optimisation, unpacking the acceptance stack transaction by transaction, modelling the impact of the RBA reforms on real transaction data, and identifying where interchange fees, Australia's scheme-fee structure and pricing model are quietly costing margin. It is the same evidence-led approach we bring across our payments consulting practice, from payments orchestration strategy and payments operations through to cross-border payments and payments compliance and risk.
None of this requires a new platform or a renegotiation to begin. Most organisations start simply, by unbundling one month of acceptance cost into its real components and seeing, often for the first time, where the margin they do not see is actually going.
If your payments cost is still a single number on a slide, that number is hiding the decisions that would actually reduce it.
Aftab Khan
Head of Payments Consulting
On a mission to help change the world, one experience at a time
I collaborate with progressive businesses to deliver successful business outcomes in 4 key areas: Increasing Revenues, Improving Operational Efficiencies, Regulatory Compliance and Elevating Customer Experience, leading to sustainable business growth.
Comments