IE 11 Not Supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

Missouri’s Fraud Finder App Targets SNAP Benefits Abuse

The data tool and interface, which was built in-house to flag crime and misuse, has saved the state millions and ensures benefits go to those in need. Created with federal funding, it recently earned a governor’s award.

A woman using SNAP to purchase groceries.
SNAP provides food benefits to low-income families to supplement their grocery budget so they can afford nutritious food.
USDA Courtesy Photo
What started as an idea about how to use data analytics for fraud detection has evolved into an in-house technical system that targets “sophisticated, fraudulent activity” including identity theft and application fraud in Missouri’s Supplemental Nutrition Assistance Program (SNAP).

Backed by a $415,025 U.S. Department of Agriculture (USDA) grant, the state’s Fraud Finder Data Analytic Application has launched more than 1,000 investigations, uncovered $1.4 million in fraudulent claims, led to 37 criminal prosecutions and saved $2 million by the end of 2023.

The project, from the Missouri Department of Social Services’ (MDSS) Welfare Investigation Unit (WIU), earned a 2024 Governor’s Award for Quality and Productivity in June, and was recognized as a model of innovation in state operations. That honor capped a roughly eight-year journey for the application, which was sparked by a 2017 audit of how data analytics were used to support SNAP program integrity, and stood up in 2022.

Kerry Howren manages the state Information Technology Services Division (ITSD) applications team that directly supports the MDSS staff. This includes Chief of Investigations Dana Carrington and Medicaid Audit and Compliance Director Richard Ferrari.

“Dana came to me and said they want to start [investigating] deeper fraud that they suspected exists within the SNAP program,” Howren said. “They were given reports of fraud from various sources, like a citizen or from an internal employee. So, we sat down with them, had them step us through which systems they were looking at, which screens they were looking at, which pieces of information they were using. And we went back to our little laboratory and started punching the numbers and trying to figure out how to automate that, because that’s what we do: We automate things.”

To activate its data in a full system, the WIU partnered with the ITSD applications team. The resulting Fraud Finder combines five modules that process tens of thousands of lines of data, producing reports with high-probability fraud matches. A web interface shows investigators key case details at a glance.

Its makers first applied for the USDA SNAP Fraud Framework Implementation Grant in fiscal year 2021, getting the award in 2022 after initially being turned down. The framework was created in 2018 to improve state efforts in detecting, investigating and preventing recipient fraud, according to its website. Ten state partnerships including Missouri now use it.

The state’s monthly SNAP benefits, colloquially called "food stamps," cover an average of about 323,462 households per month and covered about $1.5 billion in benefits for fiscal year 2024. This breaks down to about $397 per household and $196 per person each month, including some 272,073 children, according to MDSS’ FY2024 Annual Data Report.

“The success of our program, specifically the Fraud Finder to date, we have identified $4.8 million in fraud that we’re going to be bringing back,” Ferrari said. They have now taken some 59 “egregious” fraud participants to prosecutors.

Previously, the unit relied mostly on outside tips and frontline staff to report suspicious activity. Now, eligibility workers can send fraud referrals directly to investigators through an automated process. If a flagged case meets certain criteria, the system automatically creates a new investigation, cutting down on manual data entry and the chance for errors.

While the initial USDA grant focused on SNAP, the application’s design allows new fraud modules to be added as needs change. Currently, there are five.

“Anytime we find fraud and we stop that fraud, that’s money that we save that goes to families that need it,” Howren said. “Anytime we stop somebody from getting benefits fraudulently, that’s money that we can then turn around and actually give it to a citizen who is in need, and that’s what drives us. All of us.”
Rae D. DeShong is a Texas-based staff writer for Government Technology and a former staff writer for Industry Insider — Texas. She has worked at The Dallas Morning News and as a community college administrator.