Dark Reading | Checking for Fraud: Texas Community Bank Nips Check Fraud in the Bud
Texas National Bank learned from experience that 2022 was a big year for check fraud. The Financial Crimes Enforcement Network (FinCEN) reported that cases actually doubled from 2021 to 2022, and the nine-branch bank based in Edinburg, Texas, was one of those statistics. That year, the bank experienced an event that resulted in $143,000 worth of fraudulent checks demanding payment.
It started with a Texas National Bank customer, whose checks had been stolen in the mail and distributed across the country. The checks were then altered and submitted for payment. The altered checks cleared Texas National Bank, but the bank was alerted when a customer questioned a transaction. Although Stephen Wager, the bank’s chief risk and compliance officer, was able to submit warranty claims against the other bank involved and recover the money, that event was enough for Wager to make big changes.
“That’s when our fraud department was born overnight,” Wager says.
The bank had a department dedicated to compliance and risk, in line with the federal Bank Secrecy Act, which requires measures meant to detect money laundering and other illegal activities. But Wager’s group handled fraud cases through manual operations as they occurred, he explains.
The group had been doing pretty well in preventing check fraud up until that point, manually reviewing between 150 and 200 checks per day above $5,000. Clearly, that wasn’t going to work anymore. This is where automation came in.
The first thing Wager did was take advantage of the Positive Pay function offered by the Fiserv transaction processing system, which provides core banking services for the bank. Positive Pay helps detect check and ACH fraud by comparing transaction details against a list of authorized payments provided by the customer to their bank. In the case of a mismatch, a text is automatically sent to the customer, who can decline or allow the transaction.
Positive Pay was a good first step, but Wager was determined to develop a layered approach to what he calls the “fraud onion.” The team wanted to address fraud at every point, which would require building out a toolbox.
Next up was implementing Q6 Cyber, an e-crime intelligence platform that delivers information to the bank on known cyberthreats and actors by monitoring online communities, forums, private messaging platforms, botnets and other resources on the Dark Web. With that information, the bank could take proactive action to protect customer accounts against unauthorized access and fraud. While it has yet to catch anything, Wager says he’s satisfied that it will pay off at some point. It’s good insurance, he says.
The next task was finding a way to protect payees. While Positive Pay does a great job with the payor, it can’t protect the payee. To address that issue, Wager went back to a well he was very familiar with: Abrigo. He had used Abrigo’s BAM+ for compliance and risk management successfully at two previous banks when it was called “Banker’s Toolbox,” although he had never used its check fraud scenario capabilities because check fraud wasn’t much of a problem at that time.
Texas National Bank also had the solution in its tech stack for compliance, risk management, and fraud. Until the incidents in 2022, he hadn’t enabled BAM+’s check fraud scenario capabilities, but he began doing so when the company split its check fraud solution out into its own product last year under the name Abrigo Fraud Detection (AFD).
In a nutshell, AFD is a rules-based system driven by artificial intelligence (AI). It uses check image analysis to detect altered checks and identify high-volume check fraud patterns. For example, a foreign actor could write checks on the bank customer’s behalf. The system would compare those checks against the bank’s profile for that customer on 23 data points, including check borders, name boxes, fonts, and signatures.
At Texas National Bank, AFD currently examines every one of the roughly 1,900 checks that go through the system each day and may flag about 60 based on the rules the bank has set based on its tolerance level for loss. It will then immediately text the customer asking them to approve or deny the transaction. Within the first few months of implementing AFD, Texas National Bank prevented more than $300,000 worth of check fraud, and the number keeps rising.
. . .
To see the full article, visit Dark Reading, “Checking for Fraud: Texas Community Bank Nips Check Fraud in the Bud.”
