Why merchants need to move beyond dashboard watching and start treating payments as a commercial lever
Most ecommerce teams spend significant time optimizing marketing, refining checkout flows, and measuring conversion. But there is a moment that sits outside all of that analysis, which is when the customer has already decided to buy and the payment stack decides whether the business actually gets paid.
In most companies, payments are treated as infrastructure. A PSP dashboard gets reviewed occasionally. If revenue looks roughly right, the business moves on. The problem with that approach is that PSP dashboards tell you what happened. They do not tell you what you should change, where you are losing revenue, or how to recover it.
The businesses that consistently outperform their peers treat payments as a controllable commercial lever. They operate across three layers:
See – understand operational health
Monitor – spot problems before finance reports do
Grow – actively increase captured revenue
This article explains each of the 11 metrics in detail: what they measure, how merchants track them, and why they matter commercially.
SEE – Understand operational health
#1 – Approval rate
Approval rate is the percentage of attempted transactions that result in a successful authorization. It is the most direct measure of revenue capture in any payment stack. Every percentage point lost is revenue that never lands in the business, not because customers did not want to buy, but because the payment infrastructure failed to convert their intent into a completed transaction.
How merchants track it
The calculation is straightforward: approved transactions divided by total attempted transactions, multiplied by 100. The key discipline is what you exclude and how you segment the result. Refunds and voids should be excluded from total attempted transactions. The real analytical work happens in the breakdown: by acquirer, by MID, by card brand, by BIN range, by issuing country, by channel, and by time period.
A blended approval rate of 91% can conceal a 78% rate from a single issuer dragging the average down. The commercial value is in the breakdown, not the headline number.
Why it matters
On a merchant processing 250,000 transactions per month at an average value of €20, each 1% improvement in approval rate equals approximately €50,000 per month in additional captured revenue. That figure is from no new marketing spend, no product change, and no new customer acquisition.
A merchant with a blended 88% approval rate across two acquirers may have one running at 93% and another at 83% for identical BIN ranges. Routing that underperforming traffic is often a configuration change that takes a few hours and recovers hundreds of thousands of pounds annually. This is why approval rate at granular level is the foundation of every routing optimisation decision.
#2 – Decline Intelligence
Decline reason analysis classifies every failed transaction by the response code returned by the issuer and groups those codes into actionable categories. The critical distinction is between soft declines, which are temporary and recoverable, and hard declines, which are permanent and should not be retried.
How merchants track it
Transactions are grouped by response code, and the resulting distribution is tracked week-over-week. Soft declines include codes like 05 (Do Not Honour), 51 (Insufficient Funds), 65 (Exceeds Limit), and 01 (Refer to Issuer). Hard declines include 04 (Pick Up Card), 14 (Invalid Card), 41 (Lost Card), 43 (Stolen Card), and 46 (Closed Account). The goal is three action categories: retry later, authenticate, and stop.
Why it matters
Most merchants treat all declines identically. The result is twofold: recoverable revenue is written off because soft declines are not retried intelligently, and issuer relationships are damaged because hard declines are retried repeatedly. Each failed retry on a hard decline signals to the issuer that you are a risky merchant.
For a merchant with 15,000 monthly declines, correctly identifying the 60% that are soft and applying targeted retry logic through a different acquirer after an appropriate delay can recover 15-22% of those transactions. At an average value of €25, that is €34,000-€49,000 per month in revenue that most merchants simply write off.
#3 – Payment Method Performance
Payment method performance moves beyond simple usage tracking (what percentage of customers use Visa?) to actual commercial performance measurement (which method converts best for this specific customer base?). Each card brand and type carries distinct approval rates, average basket sizes, fraud exposure, and cost profiles.
How merchants track it
For each card brand and card type combination, the key figures are approval rate, volume share, average transaction value, fraud rate, and refund rate. The true measure of revenue contribution per method is approval rate multiplied by average transaction value, which gives revenue yield per attempt. E-wallet methods (Apple Pay, Google Pay) should be tracked separately, as they typically show 2-4% higher approval than manual card entry due to built-in tokenisation and biometric authentication.
Why it matters
Amex typically carries a lower approval rate (85-88%) versus Visa Debit (93-95%), but often drives 20-40% higher average basket sizes. The correct measure is approval rate multiplied by average transaction value, which gives revenue yield per attempt by method. A merchant that surfaces Apple Pay prominently on mobile checkout typically sees a 2-4% lift in approval rate on that channel at no additional processing cost.
A merchant that understands Amex drives 30% higher average basket size at 87% approval can make an informed decision on whether to actively promote Amex, particularly if they can negotiate OptBlue rates. These are commercial decisions, not payment operations decisions, and they require payment data to make correctly.
MONITOR – Early Warning
Chargebacks are formal disputes raised by a cardholder’s bank with the card scheme after a customer contests a transaction. Unlike refunds, which are merchant-initiated, chargebacks carry per-case fees, consume operational resource to contest, and are counted against scheme-level thresholds. Breaching Visa’s VAMP programme or Mastercard’s ECM programme triggers monitoring fees and can escalate to merchant termination.
How merchants track it
The chargeback rate is total chargebacks in a month divided by total transactions in that month, expressed as a percentage. Visa’s VAMP rate combines TC40 fraud alert count and chargeback count, divided by total Visa transactions. Mastercard’s ECM rate is total chargebacks divided by total Mastercard transactions. Beyond the rate, reason code categorisation is essential: chargebacks grouped into fraud, service, processing, and authorisation categories reveal where to focus operational resources, and win rate on contested cases shows how effectively the evidence is being managed.
Why it matters
The blended chargeback rate hides the cause. Fraud disputes, service disputes, and friendly fraud each require a completely different operational response. Fraud chargebacks require tooling and authentication improvements. Service disputes, which represent around 31% of volume for many merchants, typically point to fulfilment delays or product description gaps that are cheaper to fix operationally than with any fraud tool. Treating them identically wastes both money and effort.
#5 – TC40 / SAFE Fraud Alert Monitoring
TC40 (Visa) and SAFE (Mastercard) are scheme-level fraud reporting mechanisms through which issuing banks report transactions their cardholders have identified as fraudulent. These reports usually arrive 3-7 days before a formal chargeback is raised, making them the earliest actionable fraud signal available to a merchant. TC40 count feeds directly into Visa’s VAMP calculation, meaning rising TC40 rates predict VAMP threshold breach before chargeback data confirms it.
How merchants track it
TC40 rate is TC40 count divided by total Visa transactions, expressed as a percentage. SAFE rate follows the same logic for Mastercard. VAMP contribution is modelled by combining TC40 count with chargeback count divided by total settled Visa transactions. The key analytical layer is segmentation by BIN range and issuing country: concentrated fraud in specific BIN ranges almost always indicates an active carding attack that can be blocked with targeted velocity rules before it escalates.
Why it matters
The blended chargeback rate hides the cause. Fraud disputes, service disputes, and friendly fraud each require a completely different operational response. Fraud chargebacks require tooling and authentication improvements. Service disputes, which represent around 31% of volume for many merchants, typically point to fulfilment delays or product description gaps that are cheaper to fix operationally than with any fraud tool. Treating them identically wastes both money and effort.
#6 – Refund Behavior
Refund behavior monitoring tracks the volume, value, timing, and categorization of merchant-initiated refunds. Refunds are distinct from chargebacks because they are voluntary credits, but elevated or changing refund patterns are a leading indicator of product, fulfilment, or customer service problems that will generate chargebacks if left unaddressed. Payment data surfaces these issues weeks before support teams or finance reports confirm them.
How merchants track it
The refund rate is refund count divided by total transaction count, multiplied by 100. The refund value rate is total refund amount divided by total transaction amount. Average time to refund tracks the days between authorization date and refund date. Partial versus full refund split identifies whether customers are accepting partial resolution or demanding full reversal. Week-over-week and month-over-month trend tracking identifies escalating patterns before they reach acquirer risk levels.
Why it matters
A rising refund rate is almost always a signal of an upstream operational problem: a product that does not match its description, a fulfilment delay, a customer service failure, or a seasonal quality issue. These problems are cheaper to fix at source than to manage through the refund and chargeback process.
A merchant whose refund rate climbs from 2% to 5% over 60 days is losing revenue twice, on the original sale and on refund processing cost, while simultaneously signalling instability to their acquirer. Identifying whether the pattern is concentrated in a specific product line, geography, or channel allows targeted remediation rather than a broad operational response.
#7 – Cross-Border Payment Performance
Cross-border performance measures the gap in approval rates between domestically-issued cards and internationally-issued cards. This gap is one of the largest sources of silent revenue loss for merchants selling internationally, and the majority of it is a routing problem, not a demand problem. Cross-border transactions face additional friction from issuer risk scoring, scheme surcharges, and authentication requirements, most of which is addressable through intelligent acquirer routing.
How merchants track it
The cross-border approval rate is approved cross-border transactions divided by total cross-border transactions. A transaction is classified as cross-border when the issuing country does not match the acquiring country. The cross-border gap is domestic approval rate minus cross-border approval rate. Segmenting by issuing country identifies which markets have the largest gap, and therefore where a local acquiring relationship would deliver the greatest commercial return.
Why it matters
A 9% gap between domestic and cross-border approval rates does not mean 9% of international customers are fraudulent. It means the payment routing is suboptimal. Local acquiring, processing through an acquirer in the cardholder’s country, typically lifts approval rates by 4-8% on that traffic.
A merchant generating 20% of revenue from US customers through a UK acquirer, with a 14% cross-border gap on that traffic, has a clear business case for a US acquiring relationship. On €5 million annual revenue, 20% US traffic equals €1 million, and a 10% approval uplift equals €100,000 per year in additional captured revenue from a single routing change.
GROW – Revenue Optimization
#8 – Recoverable Revenue
Recoverable revenue quantifies in monetary terms the additional revenue a merchant could capture without acquiring a single new customer, purely by optimizing the payment stack they already have. It combines retry recovery potential (soft-declined transactions that could succeed on a smarter second attempt) and routing recovery potential (transactions declined on one acquirer that would statistically have been approved by another).
How merchants track it
Retry recovery is calculated as soft-declined transactions multiplied by the estimated recovery probability per decline code, based on historical retry success rates by code and acquirer pairing. Routing recovery is calculated as transactions declined on one acquirer multiplied by the estimated approval probability on an alternative, based on BIN-to-acquirer historical approval comparisons. The total recoverable revenue figure combines both, giving a forward-looking monthly opportunity figure.
Why it matters
Recoverable revenue puts a pound figure on the opportunity cost of suboptimal payment infrastructure. On a mid-market merchant processing 250,000 transactions per month, this commonly ranges from £50,000 to £200,000 per month. The figure is directly attributable to specific, fixable routing and retry decisions, making it the primary commercial justification for payment optimisation investment.
A merchant with 8,000 soft declines per month at an average value of €30 has €240,000 of declined transaction value in play. A 15% smart retry recovery rate recovers €36,000 per month, or €432,000 annualized. Adding routing recovery potential typically pushes the annual figure above €500,000 from two operational changes.
#9 – Recurring Revenue Leakage
Recurring revenue leakage measures the decline rate on recurring and subscription transactions versus one-time payments, and quantifies the revenue lost to failed renewals. Failed recurring payments represent a distinct failure mode from general declines: customers have not cancelled, they have not lost interest, and their lifetime value is proven. They are being lost purely because of payment infrastructure failures that are largely preventable.
How merchants track it
The recurring decline rate is declined recurring transactions divided by total recurring transactions, filtered by the recurring transaction flag. The one-time decline rate is the equivalent figure for non-recurring payments. The gap between the two isolates the excess failure rate attributable to recurring-specific issues, primarily expired card credentials and timing-sensitive insufficient funds declines. Revenue leakage is calculated as declined recurring transactions multiplied by average recurring transaction value.
The decline reason breakdown is critical here. Code 54 (Expired Card) typically accounts for 30-40% of recurring failures and is almost entirely preventable through network tokenisation or account updater services. Code 51 (Insufficient Funds) is timing-sensitive and recoverable with an optimised retry schedule.
Why it matters
A merchant with £127,000 per month in failed subscription renewals, with 34% attributable to expired cards, has £43,000 per month in leakage that is directly addressable by implementing account updater. At near-zero marginal cost, this is one of the highest-ROI payment optimization actions available. The customers are still active. Their payment credentials simply need refreshing.
Neither expired card recovery nor funds timing optimization requires acquiring a new customer, renegotiating a contract, or any customer-facing action. They are infrastructure fixes with immediate revenue impact.
#10 – BIN-Level Routing Intelligence
BIN-level routing intelligence identifies the specific card ranges where different acquirers produce meaningfully different approval rates on identical traffic. Rather than comparing overall acquirer performance, which masks BIN-level variance, this metric surfaces the individual routing decisions that have the highest commercial impact. It is the analytical foundation of intelligent payment routing.
How merchants track it
For each BIN range, approval rate is calculated per acquirer and gaps of more than 5 percentage points between the best and worst performing acquirer are flagged. The routing opportunity is the transaction volume currently on the lower-performing acquirer multiplied by the approval rate gap, giving estimated additional approvals. Revenue opportunity is additional approvals multiplied by the average transaction value for that BIN range. BIN gaps are ranked by revenue opportunity to prioritise the highest-value routing changes first.
Why it matters
Blended acquirer approval rates hide BIN-level variance that can be 5-10 percentage points on the same traffic. Barclays-issued Visa cards may approve at 97% on one acquirer and 89% on another, not because of overall acquirer quality differences, but because of specific relationship, routing, and scheme rule differences on that issuer’s BINs. This analysis is only possible when transaction data from all acquirers is viewed simultaneously in a single platform.
Identifying 12 BIN ranges with a gap of more than 5 percentage points and routing those BINs to the better-performing acquirer generates quantifiable monthly revenue recovery. At €47,000 per month in one representative example, this is €564,000 annually from routing configuration changes that take hours to implement, with no new customer acquisition, no product change, and no commercial renegotiation.
#11 – Provider Performance Comparison
Provider performance comparison evaluates each acquirer or PSP against a consistent set of metrics: approval rate, volume, average transaction value, chargeback rate, fraud rate, cross-border approval rate, and soft decline share. Where BIN-level routing intelligence operates at the card range level, this metric operates at the provider level, giving the overall picture of how each provider performs commercially and where routing changes and contract negotiations should be directed.
How merchants track it
For each acquirer, the metrics above are calculated for identical transaction types. The critical discipline is like-for-like comparison: overall rates mask routing differences, so performance must be compared for the same BIN ranges and transaction profiles across providers. A routing matrix is built for each combination of issuing country, card brand, and value tier, identifying which acquirer has the highest approval rate for each segment. A commercial score weights approval rate performance against chargeback rate and fraud rate to produce a risk-adjusted provider ranking.
Why it matters
Having multiple PSPs only delivers value if traffic is actively routed between them based on performance data. Most merchants with multiple providers route traffic equally or on price alone. Neither approach maximises revenue. Performance data consistently shows that different acquirers excel in different scenarios: domestic versus cross-border, high-value versus low-value, specific card brands, specific issuing countries.
A merchant with three acquirers routing traffic equally may have acquirer A at 94% on domestic Visa Debit, B at 91%, and C at 88%. Routing all domestic Visa Debit to acquirer A is a single routing rule that improves blended approval on that segment by 3-6%. Across a merchant processing €10 million per month with 30% domestic Visa Debit, that is a 3-6% improvement on €3 million of transactions, or €90,000-€180,000 in additional monthly captured revenue. Provider performance data also creates negotiating leverage with underperforming acquirers.
Marketing creates demand. Product creates intent. Payments decide whether revenue exists.
When payment data sits across multiple acquirer dashboards, merchants react after problems appear. When the information is connected and structured across SEE, MONITOR, and GROW, merchants can change outcomes before revenue moves.
The 11 metrics in this framework are not payment operations metrics. They are commercial metrics that happen to live in payment data. Each one answers a question that a CFO, a Chief Revenue Officer, or a CEO should be asking:
- How much revenue is the payment stack failing to capture today?
- Where are scheme threshold breaches going to come from in the next 30 days?
- What is the quantified opportunity from routing optimization?
- How much subscription revenue is leaking through infrastructure that can be fixed this month?
These are not questions that PSP dashboards answer. They require unified data across all providers, structured intelligence, and the ability to act on what the data shows.
The objective is simple.
See what is happening
Monitor what is changing
Grow what you capture
Payments are not just processing. They are one of the few levers that can improve revenue and cost at the same time once you can actually see what they are doing.
