-
Let's be frank: 2009 was not the year that most community banks undertook groundbreaking technology projects. Rather, projects that focused on increasing efficiency or wresting cost savings out of existing technologies and contracts found themselves at the top of most CIOs' to-do list.
March 1
There aren't too many one-of-a-kind, leading edge community bank IT projects, but Union Savings Bank's integration of its video surveillance system with biometric recognition and search software and its Fiserv teller platform surely makes that list.
A new security system pairs data from 18 surveillance cameras supplied by United Alarm with facial and image recognition software from 3VR, and ties teller-station video directly to the transactions being conducted at the time. Picture this: Someone presents a bad check at a teller station in a particular branch. Once it's identified as fraudulent, investigators can isolate the video image by searching on the transaction amount or number, grab the biometric identifying characteristics of the perpetrator, and search video from all the bank's branches to see if he or she'd been in before, and under what circumstances. All that data can be packaged and given over to law enforcement for use in investigation and prosecution.
"When a fraud takes place, we knew it'd be nice to see in one snapshot what's going on, and be able to do the cross-functional searching," says Bill McNamara, svp of technology and security at $1.9 billion-asset Union Savings, which is based in Danbury, CT.
There are also many possible proactive uses of the system, including identifying fraudsters who have been in the bank to open accounts using fraudulent identities before, and using teller station facial recognition to identify VIP customers and give them expedited service.
The initial impetus for the project was to update the bank's four-year-old surveillance system because it didn't capture enough data to be truly useful in investigations. The bank also wanted to be able to monitor all locations from a single workstation in order to be able to respond to robberies or fraudulent activity quickly. They got all that, but the idea to do a custom integration into the core banking system takes the project to a new level of utility.
"Let's face it, one of the problems with fraud and security in general is everything is working in silos and is not interconnected," says Forrester Research analyst Ellen Carney. "To connect these three components here is a very powerful solution."
The custom integration of Fiserv's ClearTouch teller system to the 3VR data was made easier by the fact that each branch's journal data is maintained locally, as well as in the core system. This integration work took just a couple of months, but Union Savings Bank request was the first of a kind for Fiserv. "The novelty about this is being able to take all the information the bank has-the surveillance video, the facial recognition, and this other data on teller transactions-and having the ability to map that digital data, pull them together and get more value out of it," says Tom Knapp, Fiserv's GM for Cleartouch.
McNamara's unique title of svp of technology and security at Union may have played a role in his ability to imagine, then execute a project that integrates physical security systems with traditional bank IT. This isn't the norm at most banks, where recent Forrester Research data found that 34 percent of IT security teams have no responsibility for physical security, and only seven percent of IT security teams are fully responsible for physical security.
Return on the bank's $275,000 investment is being measured in a number of ways. First is against the efficiency and data capture goals set at the outset; the time spent investigating any given fraud incident has been reduced by 75 percent, and the amount of video surveillance captured has increase by a factor of 80. The benefits are also being drawn to the bottom line: The use of the integrated surveillance and recognition system, tied to the Fiserv teller application, an a number of new policies and enforcement actions, all helped drive fraud losses to an all-time low of .0005 percent of assets in 2009, a reduction of 90 percent from the previous year, McNamara says.
