11:30 AM
Check Imaging Systems – Data Improvements Needed to Fight Fraud
With the highly successful, near comprehensive roll-out of check imaging systems at branches, ATMs, in-clearing channels, and the exponential growth in Mobile Remote Deposit Capture (MRDC), one could assume that imaging systems are now mature and that they provide all the data necessary for fraud management. Unfortunately, that often isn't the case. According to Celent, adoption in this area has been fast: Through 2013, 90% of US financial institutions have installed branch capture. More than 60% have offered merchant capture, and more than 50% of deposit-taking ATMs were image-enabled.
While the growth in, and deployment of, check imaging equipment and processes are indeed successful, there remain important improvements in providing fraud and risk management-related data that should be addressed. Financial institutions should consider combining fraud analysis using image data with traditional fraud analysis using account data and transactional activity.
[For more on this topic, check out: Are Mobile Payment Apps a Boon for Both Customers & Fraudsters?]
For mobile RDC, we should even include device data in the equation, but that discussion will have to wait for a future post. Simply put, financial institutions need to prioritize their image vendor capabilities to ensure that they are provided the relevant check data in order to combine image analysis, transaction analysis, account analysis, and device analysis.
By combining these four data sources, financial institutions can improve detection, reduce false positives, and successfully manage the risks inherent in the switch from physical to image clearing. Context is critical in deposit account fraud detection. Knowing the context of the image along with the account activity history (average balances, average deposit, uncollected balances, etc.) is important. Including reference data context (address changes, check orders, overdrafts, RDI’s) provide a far richer and more predictive fraud detection modeling opportunity.
There are capabilities available to financial institutions that they are probably overlooking and not incorporating in check imaging systems. For example, does your imaging provider provide a confidence level for image quality? This is important to have clear awareness of, because image quality problems can be intentional to hide or mask alterations to the check.
Does your imaging provider offer the ability to identify whether the ABA fraction on the check contains information consistent with the check's MICR line? This feature is important because in simple counterfeit detection, it would mean that fraudsters had to do the extra work to ensure their counterfeits are accurate. This and similar consistency checks provide baseline detection. Today, no one seems to be checking this, so the fraudsters generally feel that they don’t have to bother with getting this right.
It is important that your system captures the data for the payer name on the printed check. With this information you can determine if the name printed on the check and the name on the account receiving the deposit match. Is kiting a possibility on deposited checks where the payer, payee, and depositor are the same? Yes, it is.
Does your imaging provider indicate whether the pictures of the front and the back of the deposited check are from the same size check stock? Looking for restrictive endorsements or requiring them on the back of the check is good, but easy to bypass by taking a picture of the back of a different check.
Also, are you able to tell what type of endorsement is on the check? Is there a stamped endorsement on a personal account or a handwritten endorsement on a business account? Not all endorsements are the same. Normal and customary patterns are important detection tools.
There are significant improvements that can and should be made to image capture systems in order to support fraud risk beyond the clearing and settlement data currently captured. In the final analysis, you can’t manage what you don’t measure and you certainly can’t model what you don’t capture.
Wesley Wilhelm (Wes) has more than 30 years of experience in banking and consulting to the financial services industry, with extensive knowledge of fraud management, payments, and retail banking technology and operations. He has held numerous management positions in risk and ... View Full Bio