Balancing Cost and Approval Rates

Balancing Payment Cost and Approval Rates: How Much More Should a Better Route Be Worth?

A routing dashboard can make two teams look simultaneously correct.

Finance highlights one column:

processing cost

Payments highlights another:

approval rate

The cheapest route is not the one with the strongest approval performance.

So which team wins?

Neither—not yet.

Cost in basis points and approval in percentage points measure different things. Until both are translated into merchant contribution, choosing between them is largely a debate about which metric deserves priority.

The economic question is more precise:

If a higher-cost route completes more legitimate orders, how much additional payment cost can the merchant rationally pay for that improvement?

Or from the opposite direction:

If a lower-cost route saves money but completes fewer profitable payments, how much performance can deteriorate before the saving disappears?

Those questions can be solved.

The answer is merchant-specific because it depends on:

•transaction value;

•merchant contribution;

•incremental legitimate payment completion;

•payment cost;

•fraud/risk difference;

•the transaction cohort being compared.

There is no universal rule that one approval-rate point is worth 10, 20, or 50 basis points.

First, put both routes on the same denominator

A pricing comparison might say:

Route A costs 1.40%.

A performance comparison might say:

Route B approves 1 percentage point more.

Those values still cannot be compared directly.

Start with a defined population such as:

1,000 comparable eligible payment attempts

Then model what happens to those same 1,000 attempts through each route.

The brief deliberately uses a per-1,000-attempt framework because it gives pricing and performance a shared denominator.

The resulting comparison can include:

Attempts

→ approvals

→ completed legitimate orders

→ merchant contribution

− payment cost

− incremental risk cost

= net contribution

Now the route decision is expressed in money rather than competing rate metrics.

Authorization approval is not the final merchant outcome

Define:

Authorization approval rate

Approved authorizations

÷ authorization attempts

This is an important payments metric.

It is not automatically the same as:

Completed-payment rate

Completed successful payments

÷ eligible payment attempts

And neither necessarily equals:

Completed-order rate

Paid completed orders

÷ relevant checkout or payment-intent cohort

A transaction can receive authorization and still fail later in the flow.

Likewise, payment attempts may represent retries or other behavior that makes a simple approval comparison misleading.

For this article, approval is an input.

The economic destination is incremental legitimate completed orders.

Gross approval is not legitimate approval

Suppose a route approves more transactions.

That sounds good until the merchant asks:

Which transactions?

An approval increase could contain:

•additional legitimate customers;

•additional fraudulent transactions;

•different customer or issuer mix;

•transactions that another route would have recovered anyway.

Only the first category necessarily creates additional legitimate merchant value.

That is why:

gross approval uplift

must be separated from:

legitimate incremental approval/completion uplift.

The latter is what can justify paying more.

Do not compare unmatched cohorts

Historical PSP dashboards often tempt merchants into comparisons such as:

Route A: 92.8% approval

Route B: 95.1% approval

and then conclude that Route B is 2.3 percentage points better.

That conclusion may be wrong.

Perhaps Route A receives:

•more international cards;

•higher-risk transactions;

•more first-time customers;

•larger tickets.

Perhaps Route B receives:

•domestic returning customers;

•easier issuer geographies;

•lower-risk traffic.

The brief requires provider-performance comparisons to control where possible for issuer geography, card mix, transaction value, customer segment, time, fraud rules, retries, authentication, and acquiring setup.

The strongest analysis uses:

the same eligible transaction cohort assigned between routes under a controlled or otherwise defensible design.

A practical cohort definition

Before calculating break-even economics, specify:

Country/market

Issuer geography

Currency

Card/payment product

Transaction-value band

New versus returning customer

Authentication treatment

Risk segment

Recurring versus one-off payment

Time period

Retry/fallback eligibility

The exact controls depend on the merchant.

The point is not to create a statistically perfect model for every routing decision.

It is to prevent a known transaction-mix difference from being mistaken for route performance.

Percentage points are not percentage increases

If approval rises from:

94% to 95%

the increase is:

1 percentage point

The relative percentage increase is approximately:

1.06%

because:

(95% − 94%) ÷ 94% ≈ 1.06%

Those statements mean different things.

Payment teams should generally express absolute approval differences in percentage points when building this type of break-even analysis.

That avoids phrases such as:

“approval improved 1%”

which can mean two different calculations.

Price the processing difference

Suppose two routes have effective processing costs of:

Route A: 1.40%

Route B: 1.65%

for the same defined transaction cohort.

The nominal premium is:

25 basis points

because:

1.65% − 1.40% = 0.25 percentage points = 25 bps

But the merchant should calculate the actual payment cost under each route rather than stopping at the basis-point difference.

If the route approves more transactions, the higher-cost route may also process more successful volume.

Therefore:

premium bps × baseline volume

can understate the incremental processing cost.

Calculate actual modeled payment cost under both paths.

Value completed orders using contribution—not revenue

Suppose a newly completed order has a value of $100.

It is usually wrong to say:

“One incremental approval is worth $100.”

The merchant may incur:

•cost of goods;

•service delivery;

•fulfillment;

•other variable expense.

Use an approved merchant contribution measure instead.

For example:

$100 order value

with:

40% contribution before payment cost

gives:

$40 contribution per completed legitimate order

for the purpose of this illustrative model.

That $40 is not a universal margin assumption.

It should be replaced with the merchant's own economics.

The Cost vs Approval Break-Even Model

For a defined comparable cohort, calculate each route independently.

Route A

Eligible attempts

×

authorization approval rate

×

post-approval legitimate completion rate

=

expected legitimate completed orders

Then:

expected legitimate completed orders

× contribution per order

minus:

payment processing cost

minus:

relevant route-specific risk cost

=

net payment contribution

Repeat for Route B.

Then calculate:

Net route advantage

Route B net contribution

− Route A net contribution

If positive, Route B is economically better under the modeled assumptions.

If negative, Route A is better.

If close to zero, the decision may hinge on factors not adequately captured by the point estimate, such as confidence intervals, resilience, operations, or strategic routing requirements.

Worked example: 1,000 comparable attempts

The following numbers are entirely illustrative.

Assume:

Input

Route A

Route B

Eligible authorization attempts

1,000

1,000

Average order value

$100

$100

Authorization approval

94.0%

95.0%

Post-approval legitimate completion

98.0%

98.0%

Processing cost on successful value

1.40%

1.65%

Contribution before payment cost/order

$40

$40

Incremental risk difference

$0

$0

The model deliberately assumes identical post-approval completion and risk only to isolate the approval-versus-cost calculation.

In a real merchant analysis, replace those assumptions with observed values.

Route A

Approvals:

1,000 × 94% = 940

Expected legitimate completed orders:

940 × 98% = 921.2

Contribution before payment cost:

921.2 × $40 = $36,848

Successful payment value:

921.2 × $100 = $92,120

Payment cost:

$92,120 × 1.40% = $1,289.68

Net modeled contribution:

$36,848 − $1,289.68 = $35,558.32

Route B

Approvals:

1,000 × 95% = 950

Expected legitimate completed orders:

950 × 98% = 931

Contribution before payment cost:

931 × $40 = $37,240

Successful payment value:

931 × $100 = $93,100

Payment cost:

$93,100 × 1.65% = $1,536.15

Net modeled contribution:

$37,240 − $1,536.15 = $35,703.85

Net advantage

$35,703.85 − $35,558.32 = $145.53

Under these specific assumptions, Route B produces approximately:

$145.53 more contribution per 1,000 comparable attempts

despite charging 25 basis points more.

The conclusion is not that a 1-point approval increase is universally worth 25 bps.

The conclusion is:

with these merchant economics, this particular improvement exceeds this particular processing premium.

Now calculate the break-even approval uplift

The better question is not merely whether a full 1-point uplift works.

It is:

What is the smallest uplift Route B needs before the higher price becomes worthwhile?

Using the illustrative assumptions above, the break-even Route B approval rate is approximately:

94.613%

Route A approval is:

94.000%

So the required uplift is roughly:

0.613 percentage points

Under the model:

•below about 0.613 points, Route B's processing premium is too expensive;

•above about 0.613 points, the incremental legitimate contribution exceeds the extra processing cost;

•at the threshold, the two routes are approximately economically equivalent.

This is the core output of the article.

General break-even formula

Let:

Aₐ = Route A approval rate

Aᵦ = Route B approval rate

Q = post-approval legitimate completion rate

V = average transaction/order value

G = contribution per completed legitimate order before payment cost

Cₐ = Route A payment cost rate

Cᵦ = Route B payment cost rate

Rₐ/Rᵦ = route-specific risk cost for the modeled cohort

Then:

Route contribution

N × A × Q × (G − V × C) − R

where N is the number of eligible attempts.

Break-even occurs when:

Route B contribution = Route A contribution

Solve for the unknown approval rate or unknown price premium.

The formula can be extended when:

•completion differs by route;

•fraud differs;

•fixed transaction charges differ;

•contribution differs by transaction segment.

But avoid adding dimensions merely because they are available.

A complex model with weak data is not automatically better than a simple, well-observed one.

Maximum economically rational extra processing cost

The model can also answer the inverse question:

If Route B really does produce the measured approval uplift, how much more could the merchant theoretically afford to pay?

Using the illustrative assumptions:

•Route A approval: 94%

•Route B approval: 95%

•order value: $100

•contribution/order before payment cost: $40

•identical legitimate completion: 98%

•Route A processing cost: 1.40%

•no incremental risk difference

Route B's approximate break-even processing rate is:

1.806%

That means its maximum modeled premium over Route A is approximately:

40.6 basis points

before the economic advantage disappears.

Because the example premium is only 25 bps, Route B still clears the break-even threshold.

Again, this is an illustrative calculation—not a market benchmark and not a recommendation about provider pricing.

Why the maximum premium changes with merchant margin

Now change only one variable:

contribution per completed order

A high-margin digital merchant may generate substantial value from an incremental legitimate approval.

A low-margin physical-goods merchant may generate much less.

Therefore the same:

1 percentage-point approval uplift

can justify very different processing premiums across merchants.

This is why payment teams should not copy a break-even threshold from another company.

The economically rational premium belongs to the merchant's own margin structure.

Ticket size matters too

Fixed fees and percentage fees interact differently with transaction value.

If the route cost includes:

percentage charge + fixed transaction fee

the model should calculate actual per-transaction processing expense.

For example:

1.4% + $0.20

cannot be fairly compared with:

1.6% + $0.05

using percentage cost alone across widely different ticket sizes.

Use the merchant's actual transaction-value distribution where differences are material.

The broader cluster requires segment-specific route cost rather than a static “PSP costs X%” assumption.

Approval is only valuable if it completes

Suppose Route B produces more authorizations but also more failures after authorization.

Then:

authorization uplift

overstates the economic benefit.

A stronger analysis follows:

Attempt

→ authorization

→ capture/payment completion

→ legitimate completed order

and values the difference at the final economically relevant stage.

That is why the break-even dashboard should show both:

authorization approval

and:

completed legitimate orders per 1,000 attempts.

A higher approval rate can also be the wrong approval rate

Consider two routes:

Route B approves 10 more payments per 1,000 attempts.

But four of those additional approvals later generate fraud losses.

The gross uplift is:

+10 approvals

The economically useful uplift is substantially lower.

Depending on the merchant's risk economics, those extra approvals could even destroy value.

The break-even framework therefore subtracts:

Incremental risk cost

Route B risk cost

− Route A risk cost

from the incremental legitimate contribution.

Fraud is a guardrail, not an afterthought

At minimum, route comparisons should monitor:

•measured fraud loss;

•dispute/chargeback outcomes where relevant;

•false-positive behavior where defensibly observable;

•relevant risk-tool cost.

Do not optimize:

approval rate at any cost.

Likewise, do not optimize:

fraud rate at any cost.

The useful objective is merchant contribution after both performance and risk.

A route that approves more fraudulent traffic is not “better-performing”

This distinction sounds obvious.

Dashboard design can obscure it.

If the approval KPI is calculated before fraud maturation, a permissive route can initially appear stronger.

That is why merchant reporting should distinguish:

gross approval

from:

legitimate approval/completion after an appropriate risk observation window.

The appropriate window depends on the merchant and payment risk.

Do not publish a provider approval comparison without stating how legitimate performance was defined.

Historical correlation is not causal provider uplift

Suppose historical reporting shows:

PSP X approval: 96%

PSP Y approval: 92%

That does not establish:

“moving traffic from Y to X creates four points of uplift.”

Transaction populations may differ.

Causal evidence is stronger when the merchant uses:

•controlled traffic allocation;

•well-matched cohorts;

•routing experiments;

•statistically defensible quasi-experimental analysis.

Even then, performance can change over time as issuer behavior, fraud rules, seasonality, authentication, and acquiring conditions change.

Treat performance as something to monitor—not a permanent provider constant.

Use route-specific expected cost

The direct payment cost itself may vary across transactions.

Expected route cost can depend on:

•card type;

•issuer country;

•transaction value;

•currency;

•MID;

•network;

•provider.

The brief specifically warns that a static statement such as “PSP A costs 1.5%” may be inadequate when route economics vary by transaction segment.

Therefore a production routing model may evaluate:

expected cost for this transaction segment

rather than one universal provider rate.

Eligibility comes before economics

The cheapest or highest-approval route is irrelevant if the transaction cannot legitimately use it.

Before comparing routes verify:

•MID eligibility;

•geography;

•network/payment method;

•currency;

•transaction type;

•recurring credential treatment;

•merchant category;

•regulatory restrictions;

•provider capability.

The broader cluster defines least-cost routing as choosing among eligible payment paths under predetermined constraints.

The same logic applies here.

First determine candidate routes.

Then compare their economics.

Do not turn this into naive least-cost routing

A naive routing policy says:

send every eligible transaction to the route with the lowest estimated processing cost.

That can fail when the cheaper path has worse:

•authorization;

•completion;

•latency;

•fraud economics;

•reliability;

•settlement behavior;

•downstream operational cost.

The article brief makes the distinction explicit:

least processing cost ≠ least economic cost.

But this article is not the general least-cost-routing tutorial.

Its job is narrower:

price the performance difference.

Contribution per 1,000 attempts

A useful reporting metric is:

Net contribution per 1,000 comparable attempts

Legitimate completed-order contribution

− payment cost

− included risk cost

normalized to:

1,000 eligible attempts

This metric lets payments compare route economics across tests without hiding changes behind different traffic volumes.

For example:

Metric

Route A

Route B

Eligible attempts

1,000

1,000

Approval rate

94.0%

95.0%

Legitimate completed orders

921.2

931.0

Payment cost

$1,289.68

$1,536.15

Risk adjustment

$0*

$0*

Net modeled contribution

$35,558.32

$35,703.85

Net advantage

+$145.53

*Illustrative simplifying assumption only.

The economic comparison now fits on one screen.

Add confidence, not false precision

An observed uplift of:

0.7 percentage points

is not necessarily different from a break-even threshold of:

0.6 points

if the estimate is noisy.

For routing experiments, the merchant should consider:

•sample size;

•confidence/uncertainty;

•time window;

•seasonality;

•issuer mix;

•rare-risk events.

If the performance difference is not statistically or operationally stable enough to distinguish from the break-even threshold, the correct conclusion may be:

evidence insufficient

rather than:

Route B wins by 0.1 points.

Segment-level break-even can reverse the global answer

Imagine Route B performs particularly well for:

domestic debit

but poorly for:

international credit

A global aggregate can show one average winner while each transaction segment has a different optimal path.

Calculate break-even economics by segments where material, such as:

•market;

•issuer country;

•card/product type;

•transaction-value band;

•currency;

•customer type.

The routing engine can then use the result at the level where the cost and performance differences actually exist.

Do not segment until sample size becomes meaningless.

Pricing savings and lost contribution belong together

Suppose a routing change saves:

$200,000 in annual processing expense

but reduces legitimate completion enough to lose:

$400,000 in contribution.

It is not a successful payment-cost optimization.

The brief explicitly requires gross payment-cost reductions to be distinguished from net merchant economics.

Conversely, a route can cost more in direct PSP expense and still be economically rational if its measured incremental contribution is larger.

The model does not favor “cheap” or “high approval.”

It favors the higher defensible net outcome.

Those are relevant capability categories for this article because local acquiring, routing, and risk treatment can affect the route candidates a merchant evaluates.

They do not prove that:

•a particular routing path is economically superior;

•Waterfall routing always improves merchant contribution;

•any approval uplift exceeds a particular processing premium.

Those claims require merchant-specific cohort data.

Route definition

Which acquirer, MID, or routing configuration does each cohort represent?

Cost

What payment cost actually applies under the merchant's fee schedule?

Eligibility

Which transactions can use each route?

Authorization outcome

Was the transaction approved?

Completion

Did it become a successful legitimate payment/order?

Fallback

Was the transaction attempted elsewhere afterward?

Risk

Did the route alter fraud/dispute outcomes?

Time

What measurement period was used?

Without these fields, a simple “approval before versus after” story can be misleading.

Routing waterfall creates an additional measurement issue

When a transaction can fall back from one acquiring route to another, distinguish:

First-route approval

Did the initial path approve it?

from:

Final payment completion

Did any eligible path ultimately complete the payment?

A fallback architecture can improve final completion without making every underlying route individually stronger.

For merchant economics, both views are useful:

•route-level performance explains routing quality;

•final completed-payment performance explains merchant outcome.

initial acquirer approval

or:

final result after routing/fallback.

The merchant dashboard should keep cost and performance adjacent

For each major route or controlled cohort, track:

Metric

Definition

Eligible attempts

Transactions that could genuinely use the route

Authorization approval

Approved authorizations ÷ authorization attempts

Legitimate completion

Legitimate completed payments ÷ eligible attempts

Average order value

Relevant completed-order value

Contribution/order

Merchant-approved contribution measure

Payment cost

Actual attributable processing cost

Cost per completed legitimate order

Payment cost ÷ legitimate completed orders

Fraud/risk cost

Defined risk losses included in model

Net contribution/1,000 attempts

Standardized route-economic output

Break-even approval uplift

Required uplift to justify cost premium

Observed incremental uplift

Controlled measured result

Economic headroom

Observed uplift − break-even uplift

The most important two rows are often:

break-even uplift

versus:

observed incremental uplift.

Economic headroom

Define:

Approval economic headroom

Observed incremental legitimate approval/completion uplift

− required break-even uplift

If:

headroom > 0

the higher-cost route clears the modeled economic threshold.

If:

headroom < 0

it does not.

If:

headroom ≈ 0

the decision is sensitive to modeling assumptions and measurement uncertainty.

This is more useful than saying simply:

“Route B approves better.”

Sensitivity analysis prevents one bad assumption from deciding the route

The break-even calculation depends heavily on contribution per order.

Run at least a small sensitivity analysis around major uncertain inputs.

For example:

Low contribution case

Base contribution case

High contribution case

Then recalculate the approval threshold.

Do the same if fraud-cost differences are uncertain.

The output should show:

what assumption would have to be wrong for the decision to reverse?

That is often more useful to CFOs than a single point estimate.

Operational cost enters only when it materially changes between routes

This article should not become another provider-TCO model.

But if a route requires:

•materially more fallback attempts;

•extra infrastructure;

•additional reconciliation;

•unusual operational handling;

include that difference as an incremental cost.

The Pricing Optimization cluster explicitly requires net savings to account for implementation and operating effects rather than treating gross processing-price savings as the final benefit.

Keep the principle:

include differences between the routes

rather than allocating every payment-operations expense into this calculation.

The decision rule

A higher-cost route is economically justified when:

Incremental legitimate contribution

is greater than:

Incremental payment cost

•incremental risk cost

•other material incremental route cost

Or:

**Δ legitimate contribution

− Δ payment cost

− Δ risk cost

− Δ relevant operating cost

0**

That is the entire cost-versus-approval argument in one equation.

The hard part is not the arithmetic.

It is making sure every Δ represents a causal, comparable, incremental difference.

Do not make “approval” the optimization target

A routing team tasked only with maximizing approval can rationally send nearly everything to a route that performs well but is too expensive.

A finance team tasked only with minimizing payment expense can rationally send too much traffic to a cheap route that loses profitable orders.

Both teams can optimize their assigned metric and worsen merchant economics.

The shared target should instead be:

net contribution for the eligible transaction cohort, subject to merchant risk and operational constraints.

What the break-even model changes in practice

Instead of a routing meeting ending with:

“Route B approves one point better.”

the team can say:

“For this cohort, Route B needs approximately 0.61 percentage points of incremental approval to cover its processing premium. The controlled test estimates 1.0 point of legitimate approval uplift, with no material observed risk deterioration. Under the stated contribution assumptions, the route creates approximately $146 additional contribution per 1,000 attempts.”

Or perhaps:

“Route B needs 0.9 points, while the test shows only 0.4. The extra approval is real but not valuable enough to cover the price.”

Those are decisions.

The original approval rates are merely inputs.

The routing decision should survive three tests

Comparability test

Are the routes measured on genuinely comparable eligible traffic?

If no, do not claim an approval advantage.

Incrementality test

Does the measured difference create additional legitimate completed orders rather than shifted/recovered volume that would have completed anyway?

If no, do not value the whole uplift.

Economics test

Does incremental legitimate contribution exceed incremental payment, risk, and relevant operating cost?

If no, the higher approval rate is not worth its price.

When all three tests pass, the merchant has moved beyond arguing about whether approval is “worth paying for.”

It knows how much it is worth for that specific transaction cohort.

Balancing Payment Cost and Approval Rates: How Much More Should a Better Route Be Worth?

Editorial brief interpretation

Payments teams frequently argue about cost and approval as though they are competing KPIs.

Finance sees a route that is 20 or 30 basis points cheaper.

Payments sees another route that approves more transactions.

Neither number resolves the decision.

The comparison becomes useful only when both are converted into the same merchant-economic result:

How much additional legitimate contribution does the higher-performing route produce, and is that amount greater than its additional payment and risk cost?

That is a break-even problem.

Field

Editorial decision

Merchant pricing problem

Cost and approval teams optimize different metrics without translating them into a common economic denominator.

Addressable cost

Route/provider processing cost for an eligible, comparable transaction cohort.

Merchant leverage

Ability to choose or route among economically comparable acquiring/provider paths.

Optimization lever

Route/provider allocation subject to performance and risk guardrails.

Baseline

Current route economics per comparable eligible payment attempt.

Primary performance guardrail

Legitimate completed-payment/order performance—not raw authorization approval alone.

Unique evidence anchor

A higher-cost route is rational only when its incremental legitimate contribution exceeds its additional payment and risk cost.

Editorial position

Solve the decision using a merchant-specific break-even equation.

Assumption challenged

“Higher approval is worth paying for.”

Merchant decision

Determine the minimum legitimate approval/completion lift required to justify a cost premium.

Format

Approval-value break-even model.

Original asset

Cost vs Approval Break-Even Model.

The article must distinguish authorization approval from completed order, gross approval from legitimate approval, historical correlation from provider causality, revenue from contribution, and percentage-point approval changes from percentage increases.

Proposed architecture

•The two dashboards that disagree

•Put cost and approval on one denominator

•Authorization approval is not completed-order performance

•Gross approvals are not all economically good approvals

•Normalize the transaction cohort

•Price the route difference

•Value incremental completed orders using contribution

•The Cost vs Approval Break-Even Model

•A worked 1,000-attempt calculation

•Break-even approval uplift

•Maximum rational price premium

•Fraud and risk guardrails

•Correlation versus causal route performance

•The routing decision rule

Original practical asset: Cost vs Approval Break-Even Model

Input sheet

Input

Route A

Route B

Comparable eligible attempts

Authorization approval

Legitimate post-approval completion

Legitimate completed orders

Average order value

Contribution per completed order

Percentage processing fee

Fixed payment fee

Total payment cost

Fraud/risk cost

Material incremental operating cost

Route output

Legitimate completed orders

Eligible attempts

× authorization approval

× legitimate post-approval completion

Contribution before route cost

Legitimate completed orders

× contribution per completed order

Net route contribution

Contribution before route cost

− payment cost

− risk cost

− other material incremental route cost

Net advantage

Route B net contribution

− Route A net contribution

Break-even approval uplift

Solve for Route B approval such that:

Route B net contribution = Route A net contribution

Maximum economically rational cost premium

Solve for Route B price such that:

Route B net contribution = Route A net contribution

Worked model summary

Using the illustrative example:

Output

Result

Route A approval

94.0%

Route B approval

95.0%

Route A cost rate

1.40%

Route B cost rate

1.65%

Cost premium

25 bps

Route A net contribution/1,000 attempts

$35,558.32

Route B net contribution/1,000 attempts

$35,703.85

Route B net advantage

$145.53

Break-even Route B approval

~94.613%

Required uplift vs Route A

~0.613 pp

Approx. maximum Route B cost rate at 95% approval

~1.806%

Approx. maximum premium over Route A

~40.6 bps

All values are hypothetical and exist only to demonstrate methodology.

Measurement plan

Cohort

Track:

•eligible attempts;

•market;

•issuer geography;

•payment method/card type;

•value band;

•customer segment;

•authentication treatment;

•recurring status;

•risk segment.

Performance

Track:

•authorization approval;

•successful capture/payment;

•completed paid order;

•legitimate completed order;

•fallback/retry outcome.

Economics

Track:

•average order value;

•contribution per completed order;

•actual payment cost;

•cost per completed legitimate order;

•incremental payment cost;

•net contribution per 1,000 attempts.

Risk

Track:

•fraud loss;

•dispute outcome where relevant;

•false-positive signal where measurable;

•observation/maturation period.

Decision metrics

Track:

•break-even approval uplift;

•observed incremental uplift;

•economic headroom;

•maximum economically rational processing premium.

SEO metadata

H1: Balancing Payment Cost and Approval Rates: How Much More Should a Better Route Be Worth?

Meta description: Learn how to compare payment processing cost and approval performance using a merchant-specific break-even model based on legitimate completed orders, contribution margin, risk, and route cost.

URL slug: /balancing-payment-cost-approval-rates

Secondary keywords:

•payment cost vs approval rate

•payment approval economics

•payment routing approval cost

•PSP approval rate cost

•authorization rate payment cost

•payment provider approval comparison

•payment processing cost conversion

•payment route ROI

Related entities:

•authorization rate

•payment completion

•acquiring

•payment routing

•PSP

•merchant contribution

•payment processing cost

•fraud

•local acquiring

•routing optimization

•cost per completed order

Article excerpt: A higher approval rate is not automatically worth a higher processing price. Merchants can calculate the exact approval or completion uplift needed for a higher-cost route to break even on incremental legitimate contribution.

Suggested structured data: Article and BreadcrumbList.

Internal links

Payment Cost vs Payment Conversion

Link when the reader needs the broader economic relationship between processing cost and payment completion.

Routing Payments by Cost

Link when moving from the break-even calculation to transaction-level cost-aware routing.

Local Acquiring Improves Approval Rates

Link where acquiring geography and authorization performance need deeper analysis.

Net Revenue Optimization

Link when the cost-versus-approval decision must be combined with fraud, disputes, refunds, and full merchant economics.

Least-Cost Routing Explained

Optional link when explaining why lowest eligible transaction cost is not necessarily lowest economic cost.

Authoritative source recommendations

•Merchant transaction-level routing data — primary evidence for route selection and eligible cohorts.

•Merchant authorization/payment-completion data — primary evidence for performance.

•Merchant fraud/dispute data — evidence for legitimate versus risky incremental approvals.

•Merchant finance contribution model — governing input for contribution per completed order.

•Merchant PSP/acquirer fee schedules — source for actual route cost.

•Merchant controlled-routing experiments — preferred evidence for causal performance difference.

Supporting visual: Cost vs Approval Break-Even Model

Start

1,000 COMPARABLE ELIGIBLE PAYMENT ATTEMPTS

Split into:

Route A

Cost

↓

Approval

↓

Legitimate completed orders

↓

Contribution

↓

Payment + risk cost

↓

Net contribution

Route B

Cost

↓

Approval

↓

Legitimate completed orders

↓

Contribution

↓

Payment + risk cost

↓

Net contribution

Center comparison:

Incremental legitimate contribution

minus

Incremental payment cost

minus

Incremental risk/operating cost

=

NET ROUTE ADVANTAGE

Bottom outputs:

Break-even approval uplift

Maximum rational processing premium

Economic headroom

Suggested image alt text

Cost versus approval break-even model comparing two eligible payment routes across approval, legitimate completed orders, merchant contribution, payment cost, risk cost, and net contribution per 1,000 attempts.

Proposed author and reviewers

Do not imply that any review has already occurred.

Data reviewer: [Named Payments Data Scientist]

Acquiring reviewer: [Named Acquiring Specialist]

Finance reviewer: [Named Merchant Finance Reviewer]

Recommended first-hand contribution: An approved anonymized controlled-routing analysis showing comparable attempts, cost, approval, legitimate payment completion, risk, contribution, and the calculated break-even approval threshold.

The brief specifically recommends those human contribution roles.

Pricing fact-check items

For each route verify:

•merchant/provider agreement;

•pricing effective date;

•percentage processing fee;

•fixed fee;

•underlying/pass-through components where relevant;

•cross-border/FX input if included;

•failed-attempt charging;

•refund/dispute treatment if material;

•pricing tier;

•transaction segment;

•public versus negotiated pricing.

Benchmark fact-check items

A route comparison benchmark should state:

•same merchant or external source;

•geography;

•payment method;

•transaction value;

•volume;

•pricing model;

•period;

•underlying cost scope;

•performance scope.

For this article, internal same-merchant route evidence is usually more useful than broad market pricing because the decision depends on merchant-specific performance.

The cluster grades same-merchant, same-transaction-type alternative-route evidence as stronger than unrelated market anecdotes.

Routing eligibility and cost fact-check items

Verify:

•candidate routes;

•MID/account eligibility;

•acquiring country;

•issuer/card eligibility;

•method/network;

•currency;

•transaction type;

•recurring credential support;

•merchant category;

•provider capability;

•route-specific expected cost;

•actual realized cost;

•fallback path.

Do not include an ineligible cheaper route in the break-even set.

Approval/performance fact-check items

For every approval claim specify:

•authorization attempts denominator;

•authorization approval;

•payment completion;

•completed order if available;

•legitimate/fraudulent classification methodology;

•geography;

•issuer mix;

•card/product mix;

•customer segment;

•transaction-value distribution;

•time period;

•authentication;

•retries/fallbacks;

•risk treatment;

•acquiring setup;

•sample size/uncertainty where appropriate.

Historical rate differences alone do not establish provider causality.

Fraud/risk fact-check items

Verify:

•fraud-loss definition;

•maturation period;

•transaction-count versus value denominator;

•chargeback overlap;

•false-positive methodology where used;

•risk-tool difference;

•extra approvals later determined fraudulent;

•recovered amounts.

Do not call gross incremental approvals economic upside before risk is considered.

Merchant-margin assumptions

Document:

•average order value;

•contribution definition;

•variable COGS/service cost included;

•fulfillment included/excluded;

•tax treatment;

•refunds included/excluded;

•payment cost excluded from pre-payment contribution if subtracted later;

•time period.

Do not use gross payment value as contribution.

Experiment and incrementality fact-check items

A routing experiment should document:

Eligibility rule

Which transactions could use either route?

Assignment

How was traffic allocated?

Fallback

Can a failed transaction subsequently use another route?

Primary performance metric

Authorization or final legitimate completion?

Risk observation

How long before outcomes mature?

Cost

Actual or estimated route cost?

Incrementality

Would the payment have completed elsewhere anyway?

Guardrails

Fraud, disputes, latency, technical failures, customer experience, operational cost.

The broader cluster explicitly recommends treatment/control cohorts and monitoring cost, approval/completion, fraud, latency, fallback, and net contribution for routing experiments.

Regulatory and network-rule flags

Routing eligibility can depend on provider contracts, network rules, regulation, merchant accounts, geography, and other transaction constraints.

This article provides general information and does not constitute legal, regulatory, financial, accounting, or commercial advice. Merchants should verify current provider terms, network rules, contractual commitments, and applicable regulation with qualified advisers, payment providers, acquirers, and relevant authorities.

This is the safeguard specified by the cluster brief.

Amber-flag claims

“A higher approval rate is worth paying for.”

Incomplete. Its incremental legitimate contribution must exceed incremental cost and risk.

“One approval point is worth X basis points.”

There is no universal value. It depends on merchant contribution, transaction value, completion, risk, and route pricing.

“Route B approves one point more.”

Requires comparable cohort evidence.

“A 95% route is better than a 94% route.”

Not without cost, legitimate completion, and risk economics.

“Approval increased 1%.”

Ambiguous. State percentage-point versus relative-percentage change.

“The cheaper route saves 25 bps.”

Gross processing saving is not net economic benefit if completion changes.

“Higher authorization means more revenue.”

Not automatically; authorization must translate into incremental legitimate completed orders.

“Local acquiring improves approval by X%.”

Not established by the public product page. The current page verifies dynamic rerouting functionality, not a universal merchant outcome.

Requires merchant-specific contribution and cost analysis.

Before publication:

•Verify applicable route-level/merchant payment pricing from the current fee schedule; do not invent route rates.

•Confirm how Brick blended pricing applies to the intended merchant setup.

•Confirm current local-acquiring markets and eligibility.

•Confirm transaction-level fields available for route/acquirer attribution.

•Verify Reporting API fields before claiming the article's full break-even model can be built directly from that API; the public page currently verifies transaction-report retrieval and CSV output, not every analytical field required here.

•Confirm risk/fraud outcome data required to distinguish legitimate from gross approval.

•Verify fallback and retry treatment.

•Do not publish merchant routing savings without approved cost and performance data.

•If internal merchant data is available, prefer an anonymized controlled routing study to an aggregate marketing statistic.

Cluster sameness-audit fields

Title: Balancing Payment Cost and Approval Rates: How Much More Should a Better Route Be Worth?

First two sentences: A routing dashboard can make two teams look simultaneously correct. Finance highlights processing cost while Payments highlights approval rate.

Opening device: Two internal dashboards optimizing different layers—not a Provider A/Provider B or Merchant A/Merchant B anecdote.

Merchant pricing problem: Processing cost and approval performance are compared as separate KPIs rather than as one economic decision.

Addressable cost: Incremental route/provider payment cost.

Merchant leverage: Ability to allocate comparable eligible transactions between routes.

Pricing/technical lever: Provider/acquirer routing.

Performance guardrail: Incremental legitimate completed payments plus fraud/risk.

Primary format: Approval-value break-even model.

Unique evidence anchor: The higher-cost route wins only after its incremental legitimate contribution exceeds incremental payment and risk cost.

Editorial position: Use a merchant-specific break-even equation.

Common assumption challenged: Higher approval is always worth paying for.

Merchant decision: Determine required approval uplift or maximum rational processing premium.

Original framework: Cost vs Approval Break-Even Model.

Primary metric: Net contribution per 1,000 comparable eligible attempts.

Secondary metrics: Break-even approval uplift, maximum rational processing premium, legitimate completed-order rate, payment cost per completed order, economic headroom.

Conclusion pattern: Three-test routing decision—comparability, incrementality, economics—rather than an FAQ or generic final recommendation.

CTA: For the next material routing decision, calculate the break-even approval threshold before changing traffic: use one comparable cohort, actual route cost, legitimate completion, merchant contribution, and risk outcomes.

Differentiation: This article owns the direct cost-versus-approval break-even equation. It does not own general cost routing, least-cost-routing mechanics, provider proposal comparison, payment-method margin optimization, or the full payment contribution framework.