With so much concern over scams in peer-to-peer payments, Visa is introducing an upgraded version of its fraud prevention system that aims to catch criminal transactions in real time.
Under the upgraded version of A2A Protect, when a customer initiates a transfer, the bank sends the details to Visa through an API. Before the money leaves the account, a direct machine-to-machine link returns a score from 1 to 99 that grades the likelihood the payment is a scam, within milliseconds. Visa sends the member bank a notice indicating the type of scam it thinks the system has detected, leaving it up to the bank to decide what to do about it.
Making Best Use of the Data
The information sharing that A2A Protect engenders may be as valuable as the real-time fraud detection. Financial institutions can get critical updates on emerging scam hot spots and coordinated fraud activity.
These additional signals can also help identify scams before authorization. Sharing this type of data has long been recommended by financial security experts, as it leads to intelligence that would be difficult for a single financial institution to develop. The efficacy of the real-time detection will depend on how banks choose to use it.
“The fraud score could prove valuable because it gives banks a broader view of suspicious activity, helping identify fraud or scam patterns that might show up at a financial institution’s doorstep,” said Suzanne Sando, Lead Analyst in Fraud Management at Javelin Strategy & Research. “This would certainly help FIs identify suspicious behaviors that are linked to emerging fraud and scam trends.
“But the effectiveness of the fraud score depends on how financial institutions choose to handle the alerts. For a fraud score like this to be truly successful, the FIs receiving the alerts will need to use them in real time to assess risk and intervene when necessary, versus just treating it as a post-transaction signal that could help with investigation.”
Test Results Are Encouraging
Visa says a UK pilot of the product has already prevented up to $460 million in losses. According to Visa, A2A Protect not only increases fraud detection by 75% in the first six months of deployment, it also scales back false positives. One European bank said it reduced fraud by 50%, and also reduced unnecessary alerts by more than 40%.
One bank in the UK pilot mapped its account numbers to card numbers, letting Visa see the users’ debit card history as well. The extra layer found a scam the transfer data would have missed. The card history showed ordinary spending by international students before the account went dormant when the students returned home. The accounts resurfaced in a year or two as a destination for transfers from people who had been scammed.