Payment Method Fees Explained

Payment Method Fees Explained: Why Cards, Bank Payments, BNPL, Wallets, and Cash Cost Merchants Differently

Method A costs 1%.

Method B costs 2.5%.

Which method is cheaper?

If both methods convert exactly the same customers into exactly the same orders, with the same refund, fraud, dispute, settlement, and operational outcomes, Method A is cheaper.

But suppose customers who cannot—or will not—use Method A can complete their purchases with Method B.

Now 1% versus 2.5% is no longer the right comparison.

The question becomes:

How much does Method B cost for each commercially valuable order it creates that the merchant otherwise would have lost?

That distinction changes how merchants should think about payment method fees.

Cards, bank payments, digital wallets, BNPL, direct debit, cash and voucher systems, and local payment methods do not merely carry different fee percentages. They involve different participants, different funding mechanisms, different failure modes, different refund and dispute rules, different customer populations, and different operational work.

The cheapest payment transaction can therefore be attached to the more expensive commercial outcome.

And the most expensive-looking method can sometimes earn its place in the checkout by converting demand the cheaper methods cannot reach.

The denominator problem in payment-method pricing

Imagine an ecommerce merchant with an average order value of $100.

Its primary payment method costs 1%.

The merchant considers adding another method priced, for this entirely hypothetical example, at 2.5%.

During a controlled experiment:

•300 completed orders use the new method;

•180 of those customers would have completed through the existing 1% method anyway;

•120 orders would not have completed without the new method.

Transaction-fee analysis says:

1% is cheaper than 2.5%.

That statement is mathematically true but commercially incomplete.

The new method processes:

300 × $100 = $30,000

At an illustrative 2.5% fee:

Method cost = $750

Without the new method, the merchant estimates that 180 of those orders would instead have used the 1% method:

Counterfactual cost = 180 × $100 × 1% = $180

So the simplified incremental payment cost created by adding the new method is:

$750 − $180 = $570

The method produces 120 incremental completed orders.

Therefore:

Cost per incremental completed order

$570 ÷ 120 = $4.75

That does not prove the method is profitable.

The merchant still has to account for:

•refunds

•fraud losses

•disputes

•failed attempts

•financing effects where relevant

•settlement and FX

•customer support

•reconciliation

•other method-specific costs

But the comparison has moved to the correct denominator.

The commercial question is no longer:

“Why should we pay 2.5% when another method costs 1%?”

It is:

“Would we spend $4.75 of incremental payment cost to acquire an additional completed order that meets our contribution-margin requirements?”

Those are very different questions.

Transaction fee, completed-order cost, and incremental-order cost are different metrics

Merchants should distinguish three measurements.

•Transaction fee

The contractual amount charged for processing a payment.

It might be:

•percentage-based

•fixed

•percentage plus fixed

•tiered

•method-specific

•bundled with other services

Useful, but incomplete.

•Payment cost per completed order

All relevant payment costs

÷ completed paid orders

Depending on the analysis, relevant costs might include:

•successful-transaction fees

•failed-attempt charges

•refund costs

•dispute costs

•fraud losses

•gateway/provider costs

•FX or settlement-related payment costs

•method-specific support and reconciliation

This tells the merchant what the payment portfolio costs to produce completed orders.

•Cost per incremental completed order

Use this only where the merchant has a defensible incrementality methodology.

Incremental method-related cost

÷ incremental completed orders attributable to the method

This asks whether the method is creating demand rather than simply moving existing customers from another payment button.

That is the metric that prevents a common mistake: rewarding a low-cost method for processing customers who would have purchased anyway while penalizing a higher-cost method that reaches customers the merchant otherwise loses.

Why payment methods cost different amounts

There is no single global “payment method fee.”

Different methods require different combinations of:

•payment infrastructure

•account or card funding

•acquiring

•network operation

•credit

•fraud management

•collection

•authentication

•settlement

•refunds

•customer support

•compliance

•merchant integration

The economic chain behind a card payment is consequently different from the chain behind a bank transfer, a financing product, or a cash voucher.

Even central-bank studies of retail-payment costs caution that results from different payment methods and different countries are difficult to compare directly when the studies use different scopes, participants, and methodologies.

The implication for a merchant is straightforward:

Do not import another merchant's method ranking into your own checkout.

Normalize the costs using your own:

•customer base

•countries

•order values

•completion rates

•refund behavior

•fraud profile

•provider contracts

•operational architecture

Card payment costs

Cards have one of the more visible payment-cost stacks.

A typical four-party card structure includes:

•cardholder

•issuer

•card network

•merchant

•merchant's acquirer or payment provider

Interchange is typically paid from the acquiring side to the issuer. Networks can separately charge scheme or network fees, while the acquirer or PSP charges the merchant for payment acceptance and associated services.

A merchant-facing card price can therefore incorporate:

**Interchange

•scheme/network costs

•acquiring or PSP economics

•fixed transaction charges

•applicable conditional charges**

Not every provider exposes those components separately.

Card cost can also vary with characteristics such as card type, transaction type, geography, network, and commercial arrangement.

That means “card fee” itself is already an average unless the merchant models the underlying portfolio.

Costs merchants should consider for cards

Do not stop at the successful authorization fee.

Include where applicable:

•fixed transaction charges

•failed authorization charges

•cross-border costs

•FX

•refund treatment

•dispute fees

•fraud losses

•representment costs

•fraud tools

•gateway charges

•operational handling

For a high-refund travel merchant, card economics can look very different from those of a low-refund digital-service merchant even when their quoted processing rate is identical.

The detailed mechanics of interchange and scheme fees belong in their dedicated articles. Here, the important point is that card pricing represents the economics of a particular multi-party payment system.

It should not become the default benchmark against which every other method is evaluated only by percentage.

Bank-payment fees

“Bank payment” is not one universal commercial model.

The category can include:

•conventional credit transfers

•instant transfers

•account-to-account checkout products

•bank redirects

•QR-based bank payments

•payment-initiation services

•domestic real-time payment systems

Potential merchant costs can include:

•provider fee

•bank or rail charges

•payment-initiation cost

•fixed transaction fee

•percentage fee

•refund or return handling

•reconciliation

•fraud operations

•engineering

•settlement-related costs

The fact that money moves over a bank rail does not prove that merchant acceptance is free.

Brazil's Pix framework illustrates why. Banco Central do Brasil's rules allow institutions to charge legal entities for specified Pix services, including sending and receiving funds, subject to the applicable rules and commercial terms.

Europe provides a different example. Under the EU's Instant Payments Regulation, PSPs cannot charge their clients more for an instant euro transfer than for a corresponding regular credit transfer.

Neither example creates a universal merchant price for “bank payments.”

A checkout provider can still provide separate commercial services around:

•initiation

•authentication

•connectivity

•reporting

•risk

•reconciliation

•acceptance

•settlement

So the merchant should ask:

What does the underlying rail cost, what does our provider add, and what operational work remains with us?

A cheap bank transaction can still create expensive operations

Suppose a bank method has a lower transaction charge than cards but creates:

•more unmatched payments

•more manual exception handling

•slower payment confirmation

•difficult refund workflows

•more customer contacts

The merchant should allocate those costs before concluding that the method is economically superior.

Conversely, if the bank method provides reliable confirmation, strong customer adoption, straightforward reconciliation, and attractive provider pricing, those benefits belong in the comparison too.

The point is not that bank payments are expensive.

It is that:

“Bank payments are cheaper than cards” is not a sufficiently precise merchant-finance conclusion.

Digital-wallet fees

“Digital wallet” is an especially dangerous category to treat as a single payment rail.

The merchant first needs to ask:

What funds the wallet transaction?

A wallet might be backed by:

•a credit card

•a debit card

•a bank account

•stored value

•a proprietary balance

•another payment mechanism

For card-backed wallets, underlying card economics may remain directly relevant.

Apple, for example, states that Apple Pay transactions are routed to card issuers for approval and that a merchant must support the payment network of the underlying card.

That is why the merchant should not automatically create an accounting category called:

wallet cost

and assume it is independent of:

card cost.

The correct decomposition may instead be:

wallet checkout experience

→ tokenized card credential

→ card network/acquirer/issuer economics

Other wallets may operate differently.

Compare wallets by funding economics

For each wallet determine:

•What is the funding source?

•Which participant charges the merchant?

•Does the merchant pay the normal underlying card or bank cost?

•Does the wallet/operator charge an additional fee?

•Does tokenization or authentication alter fraud, acceptance, or operational outcomes?

•How are refunds and disputes handled?

•Which customers use the wallet instead of another available method?

•Does the wallet create incremental completed orders?

This separates the checkout interface from the economic rail beneath it.

BNPL merchant economics

Buy now, pay later introduces another fundamentally different cost structure because the payment experience is attached to a credit product.

The CFPB currently describes the common U.S. pay-in-four model as a consumer loan generally repaid in four installments and has continued monitoring the market through data collected from major BNPL firms.

For merchant economics, that matters.

A BNPL commercial arrangement can compensate a provider not just for transmitting a payment but for functions connected to:

•consumer financing

•funding

•underwriting or credit risk

•collection

•fraud

•installment servicing

•checkout technology

The exact allocation differs by provider and product.

That is why comparing:

BNPL fee versus card fee

without considering the financing proposition is usually incomplete.

What merchants should measure with BNPL

Evaluate:

•merchant fee

•fixed charges

•average order value

•incremental checkout completion

•cannibalization of cards and other methods

•returns

•partial refunds

•disputes

•fraud

•settlement

•contribution margin

A higher BNPL fee could be rational when BNPL materially creates profitable orders.

It can also be uneconomic if customers simply migrate from a cheaper method without increasing:

•conversion

•order value

•retention

•customer access

•contribution

BNPL's commercial value must therefore be measured rather than assumed.

Do not treat provider claims about conversion or average-order-value lift as the merchant's own incrementality result.

Cash and voucher payment costs

Cash-based digital payment methods and voucher systems create yet another chain.

An ecommerce customer might:

•create an order online;

•receive a payment reference, barcode, voucher, or code;

•visit a collection point or use a supported collection channel;

•pay;

•wait for payment confirmation;

•have the merchant fulfill the order.

Potential costs can include:

•provider fee

•collection-network fee

•fixed charge

•reconciliation

•confirmation handling

•customer support

•refund operations

•settlement

•expired or unpaid voucher handling

The denominator is particularly important.

If a merchant generates 10,000 vouchers but only 4,000 become paid orders, dividing costs by 10,000 issued vouchers can give a misleading impression of method performance.

Use:

completed paid orders

where measuring completed-order cost.

If the method creates charges at issuance or during failed/unpaid attempts, allocate those costs to the orders that ultimately complete.

Cash-access methods can still have strategic value

A cash-based method may look operationally heavier than a card payment but give the merchant access to customers who:

•do not have an appropriate card;

•do not want to use it online;

•prefer cash;

•use a locally established collection network.

The appropriate question is therefore not:

“Why are we paying more for this method?”

It is:

“How much profitable demand disappears if we remove it?”

Direct debit

Direct debit often appears attractive because its headline transaction charge can be structured differently from card pricing.

But its economics include the scheme's return and refund framework.

For example, under the current SEPA Direct Debit Core scheme, a payer can request a refund of an authorized collection during an eight-week period, while separate protections apply to unauthorized collections.

That does not make SEPA Direct Debit expensive or unattractive.

It means a merchant comparing it with cards should not assume:

low transaction fee = irreversible collected revenue.

For direct debit, model:

•provider/rail fee

•mandate setup

•failed collections

•returns

•retries

•customer support

•refund behavior

•fraud or unauthorized-payment exposure

•reconciliation

•settlement timing

The method may be particularly valuable in recurring-payment models, but its economics should be measured against successful collection and retained revenue rather than initiation alone.

Local payment methods cannot be priced as one category

“Local payment method” describes relevance to a market, not one technical architecture.

One local method might be:

•a bank transfer

•a domestic wallet

•a card scheme

•a cash network

•a prepaid system

•a direct-debit product

•an instant-payment rail

•a QR payment

Their economics can therefore be completely different.

For every local method, identify:

Customer

→ method/operator

→ provider

→ underlying rail

→ settlement path

→ merchant

Then identify who charges at each relevant layer.

This prevents procurement teams from putting every alternative payment method into a spreadsheet as:

APM — X%

and treating those rates as economically comparable.

The Payment Method Cost Matrix

The merchant should normalize every method across ten dimensions.

Dimension

What to capture

Why it matters

•Percentage fee

Percentage of successful processed value

Measures value-linked transaction cost

•Fixed fee

Per attempt, transaction, collection or order

Makes ticket size important

•Failed-attempt cost

Authorization/initiation/fraud/retry charges

Separates cost per attempt from cost per success

•Refund cost

Processing fee, retained original fee, refund operations

Important in high-return businesses

•Dispute/fraud cost

Fees, losses, operational work

Can outweigh headline processing differences

•Settlement cost

Settlement fees, conversion, relevant liquidity effect

Captures post-transaction economics

•Reconciliation/support

Allocated finance and support workload

Captures operational cost

•Customer reach

Eligible customers and method preference

Shows the addressable demand

•Payment completion

Completed orders / relevant checkout opportunities

Measures commercial performance

•Incremental order contribution

Orders attributable to having the method

Separates substitution from genuine growth

The first seven dimensions describe cost.

The final three explain what the merchant is buying with that cost.

That distinction is the central idea of payment-method economics.

Add a method-level P&L

For each payment method, build a monthly view containing:

Processed payment value

Completed paid orders

Payment attempts

Then subtract or allocate:

•transaction fees

•fixed charges

•failed-attempt costs

•refund-related cost

•dispute cost

•fraud loss

•relevant FX/settlement cost

•method-specific platform charges

•directly attributable reconciliation/support cost

This produces:

Effective method cost

Total in-scope method costs

÷ successful processed value

× 100

And:

Payment cost per completed order

Total in-scope method costs

÷ completed paid orders

Those measurements should be available even when the merchant cannot estimate incrementality.

Measure incrementality separately

Incrementality requires a counterfactual.

The merchant needs an answer to:

What would this customer have done if this payment method were unavailable?

Possible outcomes include:

•used another available method;

•purchased later;

•purchased through another channel;

•abandoned the merchant entirely.

Transaction logs alone normally cannot prove which outcome would have occurred.

If 15% of customers choose a local wallet, that does not establish that the wallet created 15% of the merchant's orders.

Many of those customers may simply prefer it to a card they would otherwise have used.

Better incrementality approaches

Depending on scale and business constraints, merchants can use methods such as:

•controlled checkout experiments;

•randomized eligible-user holdouts;

•staged market launches;

•carefully designed geographic tests;

•matched cohorts where experimentation is impractical.

The method needs to be statistically and commercially appropriate.

If the merchant lacks a defensible counterfactual, report:

method adoption

and:

method completion

but do not relabel either as:

incremental orders.

The Cost per Incremental Completed Order model

Once incrementality has been estimated:

Step 1: Measure actual method cost

Include all defined method-related costs.

Step 2: Estimate counterfactual payment cost

For orders that would have moved to another method, estimate what the merchant would have paid without the evaluated method.

Step 3: Calculate incremental payment cost

Actual relevant cost with method

− relevant counterfactual payment cost

Step 4: Measure incremental completed orders

Completed orders with method portfolio

− estimated completed orders without method

Step 5: Divide

Cost per incremental completed order

Incremental method-related cost

÷ incremental completed orders

Then compare that figure with the merchant's appropriate incremental contribution economics.

A method costing $7 to create an additional completed order could be highly attractive for one merchant and economically destructive for another.

It depends on what the order is worth.

Ticket size can reverse a fee comparison

Suppose:

Method X: 1.4% + $0.30

Method Y: 1.8% with no fixed fee

Both prices are hypothetical.

On a $5 order:

Method X:

$0.37 = 7.4% effective transaction cost

Method Y:

$0.09 = 1.8%

On a $200 order:

Method X:

$3.10 = 1.55%

Method Y:

$3.60 = 1.8%

Neither method is universally cheaper.

The answer changes with order value.

This is why percentage-only comparison fails when fixed fees differ materially.

A payment portfolio should therefore be analyzed by ticket-size distribution, not merely average ticket if the merchant has substantial variance.

Refund rates can reverse the ranking too

Suppose Method A has a lower successful-transaction price.

Method B has a higher one.

But the merchant's specific contract and method rules produce materially different economics after refunds.

A high-return merchant should therefore model:

Gross processed payment cost

and:

Net retained-order payment cost

The second denominator can be more commercially meaningful.

Do not make assumptions about whether original fees are returned when a transaction is refunded. Treatment is provider- and method-specific.

Verify it contractually.

Fraud and disputes also belong in the method P&L

A method with a lower transaction price can be more expensive after:

•dispute fees

•representment

•fraud losses

•replacement goods

•manual review

•support

Another method may have a higher acceptance fee but different fraud or dispute characteristics.

Again, there is no universal winner.

The merchant needs its own observed economics.

Settlement belongs in the comparison—but carefully

Settlement can affect payment economics through:

•settlement fee

•payout fee

•currency conversion

•settlement conversion

•timing

•reconciliation requirements

Not all treasury consequences should be assigned to the payment method.

The merchant should define the scope explicitly.

For example:

Processing cost + method-related FX + method-related settlement charge

may be appropriate.

Adding the entire treasury organization's operating budget would not be.

Which payment costs can merchants negotiate?

Negotiability depends on which participant owns the charge.

Potential negotiation targets can include:

•PSP or acquiring markup

•method-provider commercial margin

•fixed transaction fee

•volume tiers

•minimum commitments

•gateway/platform charges

•selected ancillary services

•settlement commercial terms

Underlying network, scheme, regulated, or statutory costs may be less directly negotiable by an individual merchant depending on the payment structure.

For each line item ask:

Who receives this money?

Then:

Does our provider control the amount?

This is far more useful than asking the provider to “lower the overall fee.”

Questions to ask when negotiating a payment method

For each method determine:

•Is pricing based on attempts, successful payments, or settled payments?

•Is there a fixed fee as well as a percentage?

•Are failed attempts charged?

•What happens when an order is refunded?

•Are original charges returned?

•What dispute or return fees exist?

•Does international processing change the price?

•Is FX separate?

•Are settlement fees separate?

•Are there monthly minimums?

•Are volume tiers method-specific?

•Are support or reconciliation services included?

•Which underlying costs are pass-through?

•Which charges represent provider-controlled commercial pricing?

This turns a method rate into a traceable cost structure.

Do not optimize checkout routing on fee alone

Suppose the merchant has:

Method A

•lowest fee

•strong adoption among existing cardholders

Method B

•higher fee

•important local customer preference

Method C

•highest fee

•access to a financing-sensitive segment

Routing every customer toward Method A because it has the lowest transaction fee might lower the merchant's average processing rate while reducing completed orders.

Payment-method optimization should instead consider:

**Cost

•payment performance

•customer access

•contribution**

The objective is not to minimize the average number printed beside “processing fees.”

It is to maximize profitable commerce after payment costs.

A portfolio decision framework

Classify methods into four groups.

•Low cost, high incremental value

These are economically attractive methods.

Protect their reliability and customer visibility.

•Higher cost, high incremental value

These methods deserve optimization rather than automatic removal.

Investigate:

•fee negotiation

•better routing

•ticket thresholds

•provider structure

•checkout placement

•fraud optimization

The method may still be commercially valuable.

•Low cost, low incremental value

These methods may be efficient substitutes but should not receive credit for generating demand they simply capture from another payment option.

Their value may still include resilience, preference, or routing flexibility.

•High cost, low incremental value

These are the strongest candidates for:

•renegotiation

•routing changes

•narrower customer eligibility

•reduced prominence

•consolidation

•removal

But even here, confirm whether the method serves a strategically important segment or creates operational redundancy before making the final decision.

For merchants operating internationally, this analysis becomes more difficult because payment-method portfolios change by market.

The important merchant opportunity is not simply to add as many methods as possible.

It is to apply a consistent economic framework across them.

For every method, the merchant should be able to ask:

•What does it cost?

•What makes the cost vary?

•Who charges it?

•How often does payment complete?

•What happens after refunds and disputes?

•Which customers does it serve?

•How many of those orders are incremental?

•What commercial contribution remains after payment cost?

Final recommendation

Do not rank payment methods from cheapest to most expensive using one percentage column.

Rank them using the economics of the orders they actually produce.

Start with:

Effective method cost

Then calculate:

Payment cost per completed order

Then, where credible experimentation or counterfactual analysis exists:

Cost per incremental completed order

Add:

•refunds

•disputes

•fraud

•failed attempts

•settlement

•reconciliation

•customer support

•customer reach

Only then decide whether the method deserves more traffic, less traffic, renegotiation, or removal.

A 1% payment method that mostly processes customers who would have purchased anyway is inexpensive payment infrastructure.

A 2.5% method that profitably converts customers the merchant otherwise loses can be an acquisition capability.

And a 2.5% method that merely cannibalizes the 1% method without increasing profitable commerce is simply more expensive.

The percentage alone cannot tell the merchant which one it has.

The correct objective is therefore not:

Use the cheapest payment method.

It is:

Build the payment-method portfolio that produces the strongest economics per commercially valuable completed order, while preserving customer access, payment performance, operational reliability, and the capabilities the business needs.

Payment method fee FAQ

Are bank payments always cheaper than cards?

No.

Bank-payment pricing depends on the specific rail, provider, country, contract, transaction structure, and merchant operations. Some arrangements can have attractive transaction economics; others introduce provider charges or operational costs that change the result.

Compare total cost per completed order rather than category labels.

Do digital wallets add another fee on top of cards?

Not universally.

Some wallets are card-backed, so card-network, issuer, acquiring, and provider economics can remain relevant. Other wallets can use different funding mechanisms.

Determine the funding source and commercial agreement before classifying the fee.

Is BNPL too expensive if its merchant fee is higher than cards?

Not necessarily.

The merchant should determine whether BNPL creates incremental orders, changes average order value, changes refunds or risk, and produces sufficient contribution after its higher payment cost.

A higher rate with no incremental commercial value is a different proposition from a higher rate that generates profitable incremental orders.

Should merchants disable expensive local payment methods?

Only after measuring what happens without them.

If users simply migrate to another method, removal may reduce costs.

If meaningful numbers of profitable customers abandon, the apparent saving can cost more than the fee reduction.

How should a cash or voucher method be measured?

Use completed paid transactions or completed orders as the primary success denominator—not merely vouchers issued.

Include collection, processing, unpaid-voucher, support, reconciliation, refund, and settlement costs where they apply.

Which payment method should appear first at checkout?

There is no universal answer.

Placement should reflect customer relevance, completion performance, economics, and applicable legal or scheme requirements—not transaction fee alone.

This article provides general information and does not constitute legal, regulatory, financial, tax, accounting, or commercial advice. Merchants should verify current pricing, contractual terms, payment-scheme rules, refund and dispute rules, network requirements, and regulatory obligations with qualified advisers, payment providers, banks, payment-method operators, networks, and relevant authorities.

I treated the uploaded Editorial Brief 10 as the controlling specification: this article owns cross-method payment economics, not the mechanics of any one rail, and its primary denominator is cost per commercially valuable completed order rather than transaction fee alone.

The research also supports keeping the comparison deliberately non-universal. Card fees have a distinct issuer/network/acquirer stack; card-backed wallets can preserve those underlying card economics; bank-payment pricing can depend on the rail, institution, and commercial layer; direct debit can introduce refund/return economics; and BNPL introduces financing economics rather than functioning as a simple payment rail.

Editorial interpretation

Field

Decision

Merchant cost problem

Merchant ranks methods from cheapest to most expensive using the quoted transaction fee and risks suppressing methods that create commercially valuable orders.

Fee layer owned

Total merchant economics across different payment-method categories.

Who controls the cost

Varies by method: issuers/networks/acquirers/PSPs for cards; banks/rail operators/PSPs for bank payments; wallet/funding-chain participants for wallets; BNPL provider for financing economics; collection/operator/PSP for vouchers and cash methods.

Primary variable

Payment method, funding source, geography, transaction value, provider contract, completion, refund/dispute behavior, customer access, and incrementality.

Unique evidence anchor

A 2.5% method can outperform a 1% method if enough of its orders otherwise would not exist.

Editorial position

Do not remove an expensive method until its incremental demand, completion, risk, refunds, and operational costs are measured.

Assumption challenged

“The cheapest payment method should receive the most checkout traffic.”

Merchant decision

Which methods create enough incremental commercial value to justify their all-in cost.

Format

Payment-method economics portfolio.

Original asset

Payment Method Cost Matrix + Incremental Completed Order model.

Differentiation

Unlike the other pricing articles, this piece compares economic output across rails and method categories, not pricing structures within cards.

Proposed architecture

•The cheapest transaction may not produce the cheapest order

•The denominator problem in payment-method pricing

•Why different methods have different economics

•Cards

•Bank payments

•Digital wallets

•BNPL

•Cash, voucher, and prepaid methods

•Direct debit and other local methods

•Payment Method Cost Matrix

•Measuring incrementality

•Which costs merchants can influence

•Portfolio decision framework

•Final recommendation

•Scenario-based FAQ

Original framework: Payment Method Economics Portfolio

The article's distinctive asset should combine the 10-factor Payment Method Cost Matrix with a second visual quadrant:

High incremental value

Low incremental value

Lower total method cost

Protect and scale

Efficient substitute; monitor

Higher total method cost

Optimize before removing

Renegotiate, restrict, or consider removing

This gives finance and payment teams two different views:

Matrix: Where does the money go?

Portfolio: What commercial value comes back?

That keeps the asset clearly distinct from the fee-stack and pricing-model frameworks elsewhere in the cluster.

Measurement plan

Metric

Formula

Use

Effective method cost rate

Total in-scope method cost ÷ successful payment value × 100

Normalize total cost

Cost per successful payment

In-scope method cost ÷ successful payments

Compare transaction economics

Cost per completed order

In-scope payment cost ÷ completed paid orders

Commercial denominator

Attempt-to-completion rate

Completed payments ÷ payment attempts

Detect expensive failure

Refund-adjusted cost

Payment costs after refund treatment ÷ retained orders/value

Return-heavy merchants

Dispute-adjusted cost

Payment + dispute + attributable fraud cost ÷ retained orders

Risk-heavy merchants

Incremental completed orders

Experimental completed orders − counterfactual completed orders

Causal method value

Cost per incremental completed order

Incremental method cost ÷ incremental completed orders

Portfolio decision

Incremental contribution after payment cost

Incremental order contribution − incremental method cost

Commercial outcome

A critical control: do not calculate the last three as causal metrics from ordinary method-selection data alone.

SEO metadata

Field

Recommendation

Payment Method Fees Explained: Cards, Bank Payments, BNPL & Wallet Costs

H1

Payment Method Fees Explained: Why Cards, Bank Payments, BNPL, Wallets, and Cash Cost Merchants Differently

URL slug

/payment-method-fees-explained

payment method fees explained

Meta description

Learn why cards, bank payments, wallets, BNPL, direct debit and cash methods cost merchants differently—and how to compare cost per completed and incremental order.

Article excerpt

The cheapest payment method is not automatically the cheapest way to produce an order. Compare card, bank, wallet, BNPL, direct-debit and cash-method economics using total cost, completion, refunds, risk, customer access and incrementality.

Structured data

Article, BreadcrumbList, and FAQPage if FAQ is retained

Related entities

merchant payment fee, card acquiring, interchange, payment network, bank transfer, instant payment, digital wallet, BNPL, direct debit, cash voucher, local payment method, payment completion, incrementality

Internal-link plan

Suggested anchor

Destination

Boundary

how local payment methods differ by market

Local Payment Methods

Owns which methods to offer

bank-based payment mechanics

Bank-Based Payments

Owns individual bank rails

BNPL economics and installment structures

BNPL & Installments

Owns financing mechanics

cash and voucher payment flows

Cash & Voucher Payments

Owns confirmation/refund mechanics

how digital wallets work

Digital Wallets

Owns wallet architecture

interchange fees

Interchange Fees Explained

Owns issuer/card economics

card scheme fees

Card Scheme Fees Explained

Owns network fee taxonomy

Authoritative sources to retain

Reserve Bank of Australia — Interchange and Scheme Fees. Strong neutral reference for card fee ownership and the difference between issuer-side interchange, scheme fees, and merchant PSP charges.

ECB payment statistics. Useful evidence that merchants operate in markets containing multiple major payment categories rather than one universal rail; in the second half of 2025, euro-area non-cash payments included substantial card, credit-transfer, direct-debit, and e-money usage.

Banco Central do Brasil — Pix fee rules. Useful counterexample to claims that account-to-account or instant payments are inherently free to businesses.

European Commission — Instant Payments Regulation implementation. Useful for illustrating how regulatory pricing rules can apply to a rail without defining the merchant's entire acceptance-cost stack.

European Payments Council — current SDD Core rules. Primary source for refund and return characteristics that affect direct-debit merchant economics.

Apple documentation. Useful evidence that a prominent wallet can still route an underlying payment card through the card issuer/network ecosystem.

CFPB — BNPL market research. Appropriate primary government source for the structure and continuing development of the U.S. pay-in-four market.

Supporting visual

Title:

The Cheapest Payment Method Is Not Always the Cheapest Order

Use a horizontal comparison with five method categories:

Cards → Bank payments → Wallets → BNPL → Cash/local methods

Under each, show two rows:

Transaction cost

vs.

Cost per commercially valuable completed order

Then place the Payment Method Cost Matrix underneath with icons for:

•percentage fee

•fixed cost

•failure

•refunds

•disputes

•settlement

•operations

•reach

•completion

•incrementality

Alt text:

“Payment Method Cost Matrix comparing transaction fees, failures, refunds, disputes, settlement, operations, customer reach, completion and incremental order contribution across payment methods.”

Human contribution

Proposed author:

[Name] — Payment Strategy / Global Payments

Primary reviewer:

[Name] — Merchant Finance or Payments Economics

Specialist review:

[Name] — Local Payment Methods Specialist

The strongest first-party addition would be an anonymized merchant portfolio showing:

•method adoption

•effective cost

•completed-order cost

•refund/dispute-adjusted cost

•measured incrementality

That would be substantially more valuable than adding generic industry benchmarks.

Publication fact-check controls

Pricing

Do not add generic current rates for cards, wallets, bank methods, BNPL, vouchers, or direct debit.

Verify any rate used against:

•market

•provider

•merchant segment

•payment method

•ticket size

•currency

•effective date

•volume commitment

•underlying/markup treatment

The article's numeric examples should remain labeled hypothetical unless replaced by approved first-party or official pricing evidence.

Fee ownership

Method

Validate before publication

Cards

interchange → issuer; network fees → network; acquirer/PSP economics separate

Bank payment

bank/rail/provider ownership varies by implementation

Wallet

funding source and wallet/operator pricing

BNPL

provider commercial fee and financing/risk allocation

Cash/voucher

provider versus collection-network fees

Direct debit

PSP/rail fee plus applicable return/refund mechanics

Local method

method-specific operator and settlement structure

Network/interchange

Any live card rate belongs in the dedicated interchange or scheme-fee articles unless it is indispensable to a cross-method example.

Do not turn this article into another interchange schedule.

Provider/acquirer

Verify:

•transaction-success versus attempt billing

•fixed fees

•refund treatment

•dispute fees

•cross-border fees

•FX

•gateway fees

•minimums

•settlement charges

•method-specific tiers

Merchant-model assumptions

For real portfolio modeling collect:

•order value distribution

•attempts

•successful payments

•completed orders

•provider fees

•country

•customer geography

•method

•funding source where relevant

•refund rate

•dispute rate

•fraud loss

•settlement costs

•support/reconciliation cost

•measured counterfactual behavior

Regulatory

The regulatory examples in the article are jurisdiction-specific.

In particular:

•Brazil's Pix fee rules must not be generalized beyond their applicable scope.

•EU instant-payment pricing requirements do not create a universal worldwide merchant bank-payment price.

•SEPA Direct Debit refund provisions are specific to the relevant scheme and should not be generalized to every direct-debit method.

Amber-flag claims

“Bank payments are cheaper.”

Do not publish without a defined rail, market, provider and cost scope.

“Wallets cost the same as cards.”

Too broad. Funding source and commercial structure control.

“BNPL increases conversion/AOV.”

“Cash methods reach unbanked customers.”

Only make market-specific claims with suitable evidence.

“Local methods increase conversion.”

Requires merchant/market evidence. Customer availability or preference alone does not prove incremental conversion.

“Method X creates incremental orders.”

Requires a counterfactual methodology.

“Method X has lower fraud.”

Requires comparable merchant data and a precisely defined fraud metric.

•current method count and market count;

•which methods are actively available to the target merchant segment;

•pricing structure by method;

•billing on attempts versus successes;

•refund/dispute treatment;

•settlement characteristics;

•transaction-level reporting;

•whether method-level effective-cost reporting exists;

•any first-party conversion or incremental-order evidence;

•any approved merchant case demonstrating cross-method economics.

Cluster sameness-audit fields

Field

Article 10

Title

Payment Method Fees Explained

Opening device

1% method versus 2.5% method

Merchant problem

Ranking methods entirely by transaction fee

Primary fee type

Method-specific total economics

Fee owner

Varies by method

Format

Payment-method economics portfolio

Evidence anchor

Higher-fee method produces otherwise-lost orders

Editorial position

Optimize cost per commercially valuable completed order

Merchant decision

Whether each method earns its place in the portfolio

Original framework

Payment Method Cost Matrix

Secondary framework

Cost per Incremental Completed Order

Primary metric

Payment cost per completed order

Secondary metric

Cost per incremental completed order

Example

$100 order; hypothetical 1% vs 2.5% methods

Conclusion

Cheapest transaction ≠ cheapest commercial order

Unified access enables consistent portfolio measurement

FAQ

Cross-method decision scenarios

Distinct from fee-stack pillar

Does not re-explain the entire payments fee stack

Distinct from card articles

Cards are one portfolio category

Distinct from PSP Proposal

Does not normalize competing provider proposals

Distinct from Local Methods cluster

Owns economics, not which local methods a merchant should offer

The strongest differentiator here is the counterfactual denominator. Most payment-fee comparisons stop at “Method A charges less than Method B.” This piece instead asks whether Method B creates an order at all—and then calculates what the merchant actually paid to create that additional commercial outcome.