As fraudsters become more sophisticated and card fraud is ever more prevalent, fraud protection methods have developed to match them. Much of this comes down to speed of response; the most important way to stem the rising tide of fraud is to keep ahead of it—or at least keep up.

Staying on top of fraud means acting on risky-looking transactions in real time, and card issuers are using the latest technological innovations to do this, including auto-resolution, where virtual agents reach out to customers to validate whether transactions are valid or not.

Card fraud never really goes away—it just shifts around and, like water pouring out of a leaky dyke, emerges out of a different hole. The latest figures from UK payments authority APACS show that despite the impact of chip and pin, card fraud overall rose by 14% last year to £609.9m. Card-not-present fraud in transactions conducted over the phone, online or by mail order was up 13%, accounting for more than half of all card fraud.

And while counterfeit fraud losses climbed by 18%, APACS says much of this was down to fraudsters cloning UK cards for use in European countries where chip and pin has not yet been rolled out. So chip and pin may be curbing the ability of criminals to operate on the UK high street, but organised crime has found, and will continue to find, new ways to operate.

Time is of the essence because as fraudsters try to make the most of shortening windows of opportunity before card fraud is discovered, they are getting nimbler. Banks are trying to strike that difficult balancing act between stemming their losses and maximising revenues from fee-paying customers. Inevitably, cards are sometimes blocked when they shouldn’t be.

Credit that should have been extended to a perfectly good customer ends up being withdrawn because the issuer cannot effectively establish whether a transaction is fraudulent. And after all the inconvenience of unblocking or reissuing a card which is likely to be relegated to ‘back of wallet’, the customer may even consider using a different card altogether.

So what is really needed to combat evolving card fraud? Deep transaction profiling certainly helps, but detection is only part of the answer. You also need a way of getting at suspicious looking transactions right away, and either tackling the fraud by blocking the card or validating the transaction to allow the customer to carry on spending.

High probability fraud cases are relatively straightforward to deal with. If a high value transaction appears on a card in a Eastern European country for example, the day after it has been used on the UK high street, the chances are the transaction is fraudulent, and a quick call or text to the customer can verify this before it is blocked.

But what about the increasing number of slightly less suspicious-looking cases? Contact centre agents can spend all day chasing up customers over transactions with only a low probability of fraud and have to wade through dozens of perfectly acceptable transactions before they get to the bad one. Not only is that expensive and onerous on the contact centre, it’s also time consuming, and by the time the one in 20 fraudulent transactions has been picked up, further time has elapsed.

Automated virtual agent technology takes the legwork out of this process and works over multiple channels in real time to track down fraud. It uses a combination of recorded human scripts and text-to-speech synthesis to personalise the message to the customer, and reassure them that it’s not a phishing call. Many banks even use the same voice that’s in their corporate advertising. Additionally, customers don’t have any fears over the system not recognising their voice as responses are input via the keypad.

The system never sleeps or takes a break. It can jump onto the auto-resolution of suspicious transactions within 30 seconds of a transaction being identified—that’s the equivalent of it contacting you as you are leaving the shop or logging off the Web site. And with the management information which is gathered, you can analyse very clearly at what point customers are dropping out of calls, or transferring to live agents. Users of this technology tend to constantly modify the scripts, strategies and methods they use to communicate with customers, just as they would in a peopled call centre.

We hear how technology is speeding everything up. Only a few years ago card issuers were sending out letters querying potentially fraudulent transactions which would arrive three or four days later, by which time fraudsters were long gone. The impact of Faster Payments is only going to exacerbate the fraud problem for issuers, as time windows shrink further. Now it’s time to turn the potential of automation to our own advantage.

Related Articles