Article Summary: Michelle Frank, Chief Growth Officer at Penstock, explains why post-pay payment integrity programs should go beyond recovering overpayments to identify root causes, develop new audit concepts, improve pre-pay controls, and prevent repeat errors. She explores the role of AI, advanced analytics, and payment integrity expertise in uncovering missed opportunities and driving continuous improvement for health plans.

‍

A Q&A with Penstock Chief Growth Officer Michelle Frank on why recoveries alone aren’t evidence of progress

A payment integrity program can produce recoveries every month and still fail to get better.

Michelle Frank, Chief Growth Officer at Penstock, sees health plans confronting that contradiction with increasing frustration. Their post-pay vendors continue identifying overpayments, often through the same concepts they’ve been running for years. The plan pays a fee on the recovery, the underlying problem remains and another version of the same error eventually moves through payment.

The issue isn’t whether post-pay produces value. It’s whether that value stops with the recovery or helps the entire payment integrity program improve.

‍

Q. What problems are health plans bringing to you most often?

The issues haven’t changed much over the past 12 to 18 months. They’ve just gotten louder.

The industry continues investing heavily in pre-pay while innovation in post-pay stalls. Many vendors keep running established concept libraries that identify the same types of errors repeatedly.

Plans are frustrated because they’re paying fees over and over again on the same findings. The vendor recovers an overpayment, but little is done to understand why it occurred or prevent it from happening again.

That’s the distinction plans are beginning to recognize: a program can generate recovery activity without making meaningful progress.

‍

Q. What should post-pay deliver beyond the recovery itself?

The strongest post-pay programs help explain why the dollars were overpaid in the first place.

Your data mining—or payment integrity partner—should be consistently developing new concepts rather than relying on a stagnant concept library. They should always be looking for patterns and opportunities traditional approaches miss.

Then, they should help plans use what they find. At Penstock, our findings inform pre-pay edits and configuration changes. The goal is future cost avoidance, not an indefinite cycle of paying claims incorrectly and recovering them later.

‍

Q: If plans are investing more in pre-pay edits, shouldn't post-pay data mining naturally decrease over time?

Not necessarily. While effective pre-pay edits can prevent improper payments before they occur, no pre-pay program can capture every overpayment. Post-pay analytics plays a critical and complementary role by identifying patterns, billing behaviors, and emerging risks that may not be suitable for real-time adjudication.

Just as importantly, healthcare is constantly evolving. Coding practices, reimbursement methodologies, clinical trends, regulations, provider behavior, and emerging schemes change over time. A post-pay program that relies on the same concept library year after year will gradually lose effectiveness as new opportunities for overpayment emerge.

The most successful health plans view post-pay data mining as a source of continuous innovation. New concepts, advanced analytics, and ongoing exploration of claims data are essential to uncover previously unidentified overpayments, validate the effectiveness of existing pre-pay controls, and identify opportunities to prevent future losses. Rather than expecting post-pay activity to disappear, organizations should expect it to evolve, continuously adapting to changing medical and billing patterns to ensure every recoverable dollar is identified.

‍

Q. What prevents plans from adding new post-pay partners?

A: Health plans are often cautious about adding new post-pay recovery vendors because every additional partner introduces operational complexity. Vendor onboarding, compliance reviews, IT integration, provider abrasion concerns, audit coordination, and internal oversight all require time and resources.  

As a result, the decision is driven by whether a new partner can deliver incremental value beyond what is already being identified. Health plans are looking for vendors that bring new concepts, advanced analytics, specialized expertise, or unique recovery opportunities that complement, rather than duplicate, existing programs.

The most successful post-pay partnerships are built on innovation and differentiation. A vendor that continuously develops new audit concepts, adapts to evolving medical and billing trends, and uncovers previously unidentified overpayments can generate meaningful incremental recoveries without increasing administrative burden. In that scenario, the question shifts from "Why add another vendor?" to "How much value are we leaving on the table by not identifying these opportunities?"  

‍

Q. Where does AI fit into a post-pay program that is expected to keep improving?

There is no shortage of AI in the payment integrity market today. Health plans are no longer impressed by the label alone. The real question is: What does it help us find that we are missing today to improve our program?

AI's value here isn't automation. It's speed to insight.

AI can help analyze massive volumes of claims, provider, and clinical data to detect unusual patterns, identify relationships that traditional rules-based approaches may miss, and surface potential areas for investigation. This enables payment integrity teams to develop and test new concepts more efficiently and focus resources on the highest-value opportunities.

However, AI isn’t a replacement for payment integrity expertise. The most effective programs combine AI-driven insights with clinical, coding, and audit expertise to determine which findings represent legitimate recovery opportunities. Human experts validate the concepts, quantify the impact, and ensure the results are defensible.

From Penstock's perspective, the true value of AI is not automation. It is innovation. AI helps us identify those opportunities faster, but the real value comes from transforming those insights into new, actionable audit concepts that generate incremental savings for health plans.

‍

Q. What surprises clients once they begin working with Penstock?

The level of attention they receive. Clients expect us to find opportunities, but they’re often surprised by how collaborative we are and how invested we become in their success.

We could keep recovering the same dollars indefinitely since repeat findings still generate fees, but that's not the point. We want to help the client solve the underlying problem, share what we learn and keep looking for the opportunities others have missed by consistently delivering new concepts.

‍

From Recovery to Progress

Recovery will always be an important part of post-pay. But it shouldn’t be the final outcome. The greater value lies in what each finding teaches the plan, and what happens differently because of that knowledge.

‍