Visibility into worker performance is key for states to reduce SNAP error rates

Eligibility workers play a central role in ensuring accurate and timely benefit determinations, and their performance drives the overall SNAP error rates. They are not just charged with reducing their own errors, but with helping citizens navigate complex and ever-changing processes and policies. 

It’s a demanding job, and states need visibility into where and how errors occur. Continuous, case-level monitoring links error rates to workers and process steps, enabling states to better target training, clarify policy and prioritize technology investments.

Millions in state dollars are at risk. Under new rules and program changes, SNAP payment errors must be held under 6%. Eligibility workers have the greatest effect on SNAP error rates. Historically, states lacked tools to measure quality at the worker level. Today, using data and artificial intelligence, states and counties can gain insights into worker quality that offer growth opportunities and reduce errors.

Measuring Worker Performance is Key to Assessing Technology Improvements

States are investing in new technologies — such as AI policy chatbots — to reduce SNAP payment errors. Advancing AI technologies can be powerful but also costly. While technological improvements can help limit errors, they cannot fully prevent them. 

Eligibility workers are central to adopting and using any new technologies. They have the most significant influence on eligibility errors that place state funding at risk. Quality champions (i.e., Deming, Drucker) taught us that detecting and defining errors is critical in preventing them. Therefore, if states can’t measure worker-level quality, they may struggle to effectively demonstrate the impact of major technology investments. 

Continuous Monitoring Pinpoints Where Errors Start

Random sampling and point-in-time queries can miss high-risk cases and patterns tied to specific workers. Continuous monitoring automates review of every active case and flags risk regardless of error type. Importantly, it connects findings back to the eligibility worker, allowing managers to provide specific coaching and training.

Measuring Worker Performance Affects All Three Types of SNAP Payment Errors

Eligibility workers influence all three types of SNAP Quality Control errors:

• Administrative errors (worker action)
• Household errors (unintentional, incorrect client information)
• Intentional program Violations (deliberate fraud)

Workers apply policy and enter data, educate households on reporting requirements and are the first defense in spotting fraud indicators during the eligibility process. Automated analysis, risk scoring and targeted reviews of high-risk cases can measure worker performance. 

This reveals who is making errors, which error types workers are struggling with, and why they are occurring now. That level of insight supports program oversight — and helps leaders make smarter decisions about costly technology changes while improving program effectiveness and integrity. Yet most, if not all, states still lack such insights for all eligibility workers.

Continuous Monitoring Helps Validate Technology Adoption and Impact

Eligibility workers create administrative errors by incorrectly entering data into eligibility systems and inaccurately applying policy. Technology can help by extracting data from paper applications and verification records and auto-populating the eligibility system.

This type of Intelligent Document Processing can minimize data entry errors and maximize efficiency. AI agents can act as policy and interview guides to assist eligibility workers with interpretation and practical application of policy. Monitoring worker quality provides insight into user adoption and the effectiveness of these new technological supports.

Eligibility workers contribute to household errors by not clearly communicating to beneficiaries SNAP program rules and change-reporting requirements. This includes educating them about how new technologies may be reaching out to assist beneficiaries. When eligibility workers speed through this step or written guidance is unclear, the eligibility worker and process flaws contribute to unintentional errors. 

Technology can send clients texts, notices and prompts on reporting changes. However, clients may not act without understanding why this technology is reaching out to them. Quality monitoring can identify workers that lack sufficient skills in educating clients by detecting higher concentrations of household errors within high-risk case clusters.

Eligibility workers can miss intentional program violations by not understanding fraud indicators or speeding past warning signs. Workers are the first line of defense — interacting with the client during application and recertification, viewing verification documents and receiving alerts about potential issues, for example unreported income.

Often, eligibility workers are not well trained in fraud prevention and detection and lack interviewing skills to address fraud. Technology can help with improved alerting, data matches and AI-based prevention tools that can spot high-risk applications. This allows states to move these cases to more seasoned eligibility staff or even investigative staff for more intensive review. Continuous quality monitoring can identify which staff have the skills to tackle potentially fraudulent cases, those who failed to heed warning signs, and inside actors colluding with fraudsters.

Continuous Monitoring Measures Across a Shifting Workforce

Some states may seek to simply hire more eligibility staff to address error rates. Without continuous monitoring, this presents problems:

• Hiring more staff becomes an exercise in hope
• Lag time to hire and train workers limits short-term effectiveness
• Inexperienced eligibility workers often make more mistakes
• Most errors are structural and systemic (i.e., documentation or policy)

Staff churn makes it harder for a state to hire its way out of SNAP eligibility error problems. With the most experienced workers retiring and newer workers staying for shorter periods, the staff mix is shifting faster than ever before. Sampling and periodic data queries fail to account for these continual shifts.

Assessing risk of error for every active case, at least monthly, allows states to measure error trends from the state level down to the individual worker level. Continuous monitoring can also identify new workers that are struggling or more experienced staff with unexpected error increases. This allows managers to spot problems more quickly and take interventions to support eligibility staff in reducing errors.

Chasing more bodies is an admission that states perceive that staffing levels cannot meet current workload demands. However, if states are unlikely to get more staff, they must make current staff more productive. W. Edwards Deming said, “Improve quality, you automatically improve productivity.”

Leveraging Technology Empowers Operations Staff

With data and AI, continuous monitoring systems measure macro-level trends all the way down to micro-level worker performance. This empowers Operations at each level:

• Quality Assurance staff see the big picture to drive strategy
• Middle managers have data to lead tactical operations
• Line supervisors spot errors and adapt quickly
• Workers have insights to help them improve

Advancing technologies show real promise, but when it comes to getting benefits to people in need, the human in the loop remains critical. Technology can help take that human’s performance to new levels, reduce error rates and save states billions of dollars.

John Maynard is principal industry consultant at data and AI company SAS, and is a former state Medicaid program integrity director. With nearly 25 years in state government, he helps healthcare and government social benefit programs implement data-driven solutions to their greatest challenges.