Better data analysis helps banks rile fewer customers while trying to stop credit card fraud.
Big Data's Surprising Uses:
From Lady Gaga To CIA
(click image for larger view and for slideshow)
Have you ever charged something expensive -- and perhaps out of the ordinary for you -- only to have the transaction rejected by the bank? Not only do these "false positives" annoy consumers, they worry credit card issuers, who fear their customers might switch to a competitor's card.
Recent improvements in real-time fraud detection and streaming analytics are helping to reduce false positives, according to Doug Clare, VP for fraud solutions at FICO, the provider of analytics and decision-making products that's perhaps best known for its credit-scoring service.
For banks, managing credit card fraud is a delicate balancing act, Clare told InformationWeek in a phone interview. "If they want to, they can stop all fraud by declining every transaction, or they can prevent zero fraud by approving all of them," said Clare, a 25-year FICO veteran who's managed the company's fraud products for the last five years.
Some 9,000 banks use FICO's Falcon predictive analytics software for fraud prevention. Falcon "is kind of the original streaming analytics application," said Clare. "It looks at transactions as they fly by. It records certain characteristic information about transactions -- things like velocity, different spending types, and favorites in terms of where you shop and spend."
Falcon's neural network models allow it to predict if a particular transaction is fraudulent. It also looks for activities that are out of character for you as a customer. "If you never go to a casino, but all of a sudden you have a $500 withdrawal at a casino at two in the morning, that might be out of character for you," Clare explained.
As the economy has gradually improved over the past two to three years, FICO's bank customers have begun focusing more on making life easier for their card-carrying clientele, and that means fewer charge rejections.
"While you never have a fraud conversation without talking about preventing losses, most of the conversations now start and end with 'enhancing the customer experience' or 'improving the customer experience,'" said Clare.
This process involves finding ways to enhance the accuracy of Falcon's analytics, as well as honing the banks' communications skills with their customers. "If banks can more reliably get in touch with the consumer, and do that in real-time or near-time basis, they may be willing to approve a suspicious transaction," said Clare.
For this to happen, however, the line of communication between banks and their customers must be open and immediate, she noted. "The big banks have big call centers, and they're monitoring transactions. If something comes up that's suspicious, they'll decline it -- or they may approve it and call you," said Clare. "Everybody's gotten that phone call, 'Did you make this transaction?' Those calls are initiated typically (because) the Falcon score has indicated some level of risk."
You might find you're getting fewer of those phone calls, however, as financial institutions opt for alternative means of reaching out, including email, text message and even via mobile apps. "If you're a Bank X customer, you might have the Bank X mobile banking app, which provides a context for real-time alerts," Clare said. "If we can drive the message to the consumer that way, and try to validate the transaction, it drives down the cost of the call."
And that might be what banks find most appealing about those email, text and mobile app notifications: they cost less than having call center employees contact you via phone. And they're far less intrusive than phone calls.
As Clare sees it, these new methods help prevent fraud, reduce losses and allow banks to decline fewer transactions.
Then again, if you decide to charge a $5,000 diamond necklace while traveling in Kyrgyzstan, don't be surprised if the transaction is rejected.
6 Tools to Protect Big DataMost IT teams have their conventional databases covered in terms of security and business continuity. But as we enter the era of big data, Hadoop, and NoSQL, protection schemes need to evolve. In fact, big data could drive the next big security strategy shift.
Big Data Brings Big Security ProblemsWhy should big data be more difficult to secure? In a word, variety. But the business won’t wait to use it to predict customer behavior, find correlations across disparate data sources, predict fraud or financial risk, and more.
Join us for a roundup of the top stories on InformationWeek.com for the week of December 14, 2014. Be here for the show and for the incredible Friday Afternoon Conversation that runs beside the program.