By: Kennis Wong
In the last post, I emphasized the importance of fraud detection even after an account has been approved. If information gathered later indicates an application was fraudulent, credit issuers can still take action on the account to minimize fraud losses.
Monitoring your internal systems to find suspicious activities is one way to do it. If the account holder has unusual purchase patterns, such as spending $2000 at a dry cleaner, you may want to stop and have a closer look. But more revealing would be the bigger picture – Is the account holder developing other financial relationships? Do these other applications indicate high identity theft risk? Are there any unusual patterns across the multiple financial relationships?
The tricky part is finding the related applications. If you are looking for applications that use the same SSN, name, DOB, address and phone number, you may be missing information that helps detect fraud. Fraudsters often mutate elements of the PIIs when they use stolen identities to hide their fraudulent activity. If you link related applications together, you can then look for unusual patterns collectively. Find that the same social security number was used 10 times, with different addresses, all in the same week? Bad sign.
Individual signs may help very little. False-positives and fraud referral rates may be too high if your action is based on just one or two signs. That’s why Experian recommends using a risk-based method for minimizing fraud instead of a rule-based method. You need fraud analytics to put all signs together in a way that is predictive of identity theft.
Timeliness is the key to successful fraud account management. If the identity fraudster has already used all available credit on a credit line, then it is too late to minimize fraud and action on the account. The only benefit at that point — saving time by telling your collection department not to waste effort attempting to collect on the account.
