Subscribe / Unsubscribe Enewsletters | Login | Register

Pencil Banner

Preventing fraud the smarter way

Nurdianah Md Nur | Oct. 2, 2014
Analytics can be used to compare a customer's historical activity to his current activity and determine if the activity is suspicious, said Adam Davies of FICO.

With fraud being so prevalent, who should be responsible for fraud prevention?
A recent survey revealed that nearly half of customers performing online transactions are confident that their banks would refund them if money is stolen from their online account. However, the truth is that banks do differ on how and when they compensate customers. Most banks are keenly aware that it is in their interest to be at the forefront of fraud prevention. Fraud reduction is actually becoming an integral part of banks' customer retention strategy. Statistics have shown that poor handling of fraud is a key reason for losing a customer's trust. In fact, one study finds that 1 in 3 customers will likely close an account or use it less as a direct result of being a victim of a fraud. 

However, as fraud may happen either at the point or sale, at ATMs or online, it is necessary for all these payment channels to be safeguarded with available technologies that would help prevent fraud at all levels.

One of the best ways to do this is using predictive analytics. It is a statistical approach used by FICO where we are able to use years of historical data to create models of purchasing behaviour. We then use these models along with live data to accurately predict if a transaction looks suspicious or genuine. It is a key tool for fraud because it can be used across the portfolio of credit and debit cards, and even monitor the point of sale and ATMs.

New forms of payment such as mobile banking and contactless cards have sprouted since the beginning of this year and they have been gaining traction in certain parts of Asia. In what ways are such modes of payment increasing the risk of fraud?
With Asian consumers starting to embrace more convenient and non-traditional payment modes and channels such as mobile banking and contactless cards, the criminals will always look for ways to attack the payment system. 

There will be attempts to compromise the terminals and the ecosystems. No banking channel may be totally immune from threats of fraud. Banks look at a certain level of fraud as a 'cost of doing business', but they need to work hard to keep it at a manageable level.

For FICO, it doesn't matter how the transaction occurs -- that's the beauty of predictive analytics. It looks for oddities in payment patterns regardless of whether the purchase happened online or at the store using contactless. However, it is worth noting that using predictive analytics for fraud protection is relatively new amongst some Asian banks, particularly in markets where credit growth has been very recent and explosive.


Previous Page  1  2  3  4  Next Page 

Sign up for Computerworld eNewsletters.