As state agencies navigate increasing caseloads, evolving program requirements, and heightened scrutiny around eligibility accuracy, quality assurance has become more important than ever. New federal policies are raising the financial stakes of eligibility errors, prompting agencies to rethink how they monitor accuracy, support staff, and maintain program integrity at scale.

We sat down with PCG’s Peter Cheesman and Jay Derby to discuss the changing eligibility landscape, where agencies face the greatest risks, and how modern quality assurance solutions like PCG’s Auto.IES™ can help improve accuracy, reduce risk, and support frontline eligibility workers.

How is the eligibility landscape changing for state agencies?

Jay Derby: Some challenges never really go away. Across assistance programs, agencies are handling bigger caseloads with the same staff, and the rules keep getting more complicated. That strain has been building for years.

What’s new is the cost of getting it wrong. Accuracy has always mattered, but now it’s tied straight to a state’s budget. Under H.R.1, both SNAP and Medicaid link error rates to real financial consequences, states can owe a share of costs or lose federal funding once their error rates cross set thresholds. This is part of a broader effort to strengthen program integrity and accuracy across multiple benefit programs, with several important deadlines approaching soon.

Where do agencies typically see the greatest risk for eligibility errors and inconsistencies?

Jay Derby: Understaffing is probably the greatest risk factor for Medicaid eligibility errors or inconsistencies. Eligibility work is genuinely complex; income calculations, household composition, and cases that don’t fit the standard pattern all take time and care to get right. Skilled workers handle this every day, and high caseloads, changing rules, and limited resources can make consistent decisions harder to sustain. It’s less about people and more about making sure teams have the support, tools, and capacity they need.

The harder part is that most agencies don’t have a way to check all of it. Full-scale QA tools haven’t really existed, so the standard approach is to sample a small slice of cases after certification. That’s enough to estimate an error rate, but not to catch a specific problem before it spreads.  The most significant errors can often be the ones the sample never looks at.

What does a modern approach to eligibility quality assurance look like?

Peter Cheesman: Simply stated, a vision of modern QA would include an automated review of every case before eligibility is certified that doesn’t burden eligibility staff with any more responsibilities than they already have.  Specifically, modern QA should be continuous and proactive, rather than relying only on the review of only a small sample of cases after it’s already too late. Advances in technology can help agencies review virtually all of their preliminary eligibility decisions, identify accuracy or procedural errors, and flag cases that need additional attention before eligibility is certified. 

Additionally, QA also needs to be transparent and auditable. Workers should understand why an issue was flagged, what evidence supports the flag, and what action is needed to adjudicate it. The goal is not to replace eligibility workers; rather, it is to support them in completing a difficult job more quickly and accurately.

What insights has PCG gained from working with state agencies that helped shape the development of Auto.IES?

Peter Cheesman: One of the biggest lessons we’ve learned from our agency clients is that many are scrambling to address eligibility quality through a patchwork of partial solutions, including additional data sources, document extraction tools, updates to worker trainings, and more. While each of these approaches has proven effective, none of them can completely solve the issue of eligibility errors on their own. This, combined with the fact that eligibility errors rarely have a single cause, suggested a gap in what the market was offering states in response to H.R. 1 financial pressures. This observation really pushed us to think through how advances in technology could be combined to create a complete, automated QA service which could evaluate both structured and unstructured data, documentation, and metadata to flag any accuracy or procedural errors immediately before eligibility is certified. 

We also learned that states need quality assurance closer to the point of decision. Finding an error months later during a sample review is helpful, but preventing or correcting it before the eligibility decision is finalized is much more valuable. These insights and observations, and more, helped shaped our approach to developing Auto.IES. By employing AI, NLP, and powerful document extraction tools, Auto.IES can review the full context of a case, provide clear and evidence-based findings with policy citations, and keep the eligibility professional in control of the final decision. At the same time, it also helps agencies see broader trends, such as recurring policy issues, training needs, documentation gaps, and system configuration problems. 

At the end of the day, we wanted to build a simple solution that would make things easier for states and their eligibility workers. And that’s what we did. At its core, Auto.IES simply automates the checks of an eligibility worker’s work before it’s too late. 

Discover how Auto.IES can help your agency improve eligibility accuracy and reduce compliance risk, learn more here.