|View printer-friendly version|
|Visa Announces Improved Payment Card Fraud Detection|
Financial institutions can use the information to more reliably know which transactions to decline in real time, potentially reducing fraud by billions per year, while more confidently approving legitimate transactions to remove friction from payments for merchants and consumers alike.
"Cardholders, merchants and issuers all want to have confidence in the convenience and the security of every Visa transaction," said
Visa has increased the breadth of each account profile in the Advanced Authorization model by adding more transactional history data, along with additional neural networks to analyze that data. The account profile is a major component of the risk score assigned to a given transaction and provided to the issuer for them to make an authorization decision. The result is more robust performance and improvement of as much as 130% in detecting fraud in debit transactions and 175% for credit transactions.
The enhanced model includes additional risk indicators specific to Automated Fuel Dispensers (AFD) transactions. Visa's network now can pinpoint suspicious activity at a gas station and apply that to all transactions processed through that station. The model also uses account velocity at AFDs compared to that account's normal behavior in the score determination. This can potentially increase the effectiveness of fraud detection in this segment by as much as 266% for debit transactions and 163% in credit.
The VAA improvements offer the potential to substantially reduce fraud in both transactions where the physical Visa card is present, such as a retail store, as well as in "card not present" environments such as online shopping. The improvements are effective for both consumer and commercial accounts and transactions.
Sandra Chu, Visa Inc., +1 650-432-1217, Email: email@example.com