Case Study

Summary

To properly serve nearly 67 million Medicare beneficiaries, the Centers for Medicare & Medicaid Services (CMS) relies on modern, resilient infrastructure to process and monitor claims. In 2024, CMS’s Medicare Payment Systems Modernization (MPSM) ecosystem processed over $1 trillion in claims, accounting for 21% of the U.S.’s national health expenditure.

The MPSM environment supports dozens of applications and several Application Development Organizations (ADOs), which build and support CMS’s applications. Given the far-reaching impacts of CMS’s payment systems, it’s crucial for ADOs to have real-time visibility into system health, performance, and reliability.

As part of a broader modernization and cost optimization effort, CMS tasked our team with assisting in a migration to Datadog — a platform that offers in-depth system monitoring. The effort required transitioning monitoring for 52 applications across 7 ADOs, all without disrupting live claims processing. 

Approach

CMS’s priority was to make the transition to Datadog as easy as possible for ADOs. By reducing friction and setting ADOs up to succeed, CMS was able to lead an extremely efficient and risk-free cutover. 

We helped CMS design a scalable, zero-risk approach that allowed ADOs to plug right into Datadog. Specifically, we developed a “sidecar” monitoring solution that worked alongside CMS’s existing infrastructure. We also provided ADOs with robust documentation, guidance, and support. This approach minimized the burden on ADO teams while maintaining high reliability standards.

Outcomes

In just one month, Nava helped the majority of CMS’s seven ADOs migrate to Datadog, with full completion shortly thereafter. This also required migrating 52 total applications from across the MPSM ecosystem. Throughout the transition, we helped CMS continuously operate real-time claims processing systems with zero downtime.

With Datadog in place, CMS will enjoy more cost savings and enhanced visibility into system performance. The tool enables CMS to trace data throughout its systems, which can help surface latencies and bottlenecks. CMS will also receive real-time alerting and incident management. And finally, Datadog provides CMS with unified dashboards across applications and infrastructure.

Process

Rather than requiring ADOs to configure Datadog individually, we worked with CMS to identify ways to reduce complexity for these teams. With CMS, we decided to develop a sidecar monitoring solution that could plug directly into existing systems. Our team preconfigured the sidecar for nearly all of the information ADOs would need, including tagging resources and creating logic. This enabled the ADO teams to deploy Datadog with minimal configuration and zero downtime. The sidecar solution also offered automatic application health checks and collected application metrics like latency and CPU usage. 

To ensure consistency across ADOs and ease of use, we paired the technical solution with robust support. This included comprehensive and centralized documentation, clear migration guidance tailored to MPSM applications, and built-in deployment support.

Despite the complexity of this transition, we were able to help CMS successfully transition monitoring systems with zero downtime. 

Conclusion 

By centering the needs of ADOs, we helped CMS conduct a highly successful cutover to Datadog without disrupting critical operations. The result is a more cost-effective, reliable, and transparent monitoring platform that empowers CMS teams to quickly detect, diagnose, and resolve application issues. At a high level, this work strengthens CMS’s ability to deliver consistent, high-quality services at scale. 

This work also demonstrates Nava’s human-centered approach to engineering, ensuring critical systems remain operational while strengthening the security position to protect beneficiary data.