Card fraud cannot be fully eliminated in cross-border payments. The objective is therefore not to remove risk entirely, but to manage it intelligently and efficiently.
For merchants, overly strict risk controls may reduce fraud but often come at the cost of lower approval rates and weaker customer experience. On the other hand, overly relaxed policies may increase conversion in the short term, but can lead to higher chargebacks, financial loss, and stricter scrutiny from card schemes over time.
In practice, both extremes reduce long-term business performance.
The optimal approach is to find a balanced point between approval rate and fraud exposure, where revenue is maximised while risk remains under control.
This is the core principle of modern payment risk management: not minimising risk at all costs, but optimising overall business outcomes.
Business Risk–Return Matrix
Payment approval rate and fraud chargeback rate are two closely linked but distinct metrics in cross-border payments.
The payment journey involves multiple steps—from user acquisition to checkout completion—and friction can occur at any stage. A higher approval rate generally improves conversion, but it may also increase exposure to fraudulent transactions if controls are too loose.
Conversely, tighter fraud controls may reduce chargebacks but also reject legitimate customers, lowering revenue.
This creates a natural tension: improving approval rates while keeping fraud and chargebacks under control is essential for sustainable growth.
The most effective approach is granular, transaction-level risk decisioning, rather than applying uniform rules across all payments.
Each transaction carries different signals—such as device, IP, order value, behavioural patterns, and location—and should be evaluated individually.
This is where advanced risk systems such as those developed by Antom come into play. Antom applies a multi-layered risk framework combining real-time detection with post-transaction analysis, enabling merchants to dynamically manage fraud exposure.
During checkout, transactions are evaluated in real time and classified into three outcomes:
l Accept: low-risk transactions proceed smoothly
l Reject: high-risk transactions are blocked immediately
l Step-up verification: medium-risk transactions undergo additional authentication (e.g. 3DS)
This adaptive approach helps merchants reduce fraud while preserving approval rates and customer experience.
Beyond real-time screening, post-transaction analysis further strengthens risk intelligence by identifying emerging fraud patterns and continuously refining detection models.
Together, this creates a closed-loop risk system that continuously learns and improves over time.
Common verification methods include card details such as card number, expiry date, and CVV, but these alone are no longer sufficient in a global fraud environment.
A key additional layer is 3-D Secure (3DS), an authentication protocol involving the issuer, card scheme, and acquirer. It enables issuing banks to verify cardholder identity during online transactions.
When 3DS is triggered, customers are redirected to their bank for authentication before completing payment. If successfully verified, liability for fraud typically shifts from the merchant to the issuing bank.
This not only reduces merchant exposure to chargebacks but also strengthens transaction security in card-not-present environments.
While 3DS improves security, applying it universally is not always optimal.
Mandatory 3DS for all transactions can introduce friction into the checkout process, potentially reducing conversion rates due to additional authentication steps and page redirects.
It can also increase operational cost and negatively impact user experience, particularly in mobile-first environments.
Therefore, the key is not “whether to use 3DS”, but when and where to apply it strategically.
Modern risk systems such as those from Antom use real-time risk scoring to determine whether a transaction should be:
l Approved directly
l Blocked
l Or routed through 3DS verification
This selective approach allows merchants to reduce fraud without sacrificing conversion. In many cases, merchants adopting this model have seen significant improvements in both approval rates and fraud reduction.
3DS strategy is not universal and must be adapted to regional payment environments.
In markets such as Brazil, where authentication adoption is relatively limited, merchants often rely more heavily on real-time fraud screening combined with selective 3DS activation.
In the United States, where scheme rules and fraud monitoring are strict, 3DS usage must be carefully calibrated to avoid excessive friction and compliance risks.
In Europe, regulatory frameworks such as PSD2 and Strong Customer Authentication (SCA) require additional authentication layers, making 3DS a more embedded part of the payment journey.
This “market-specific strategy” ensures that fraud controls are aligned with both regulatory requirements and local payment behaviour.