Inova Health compresses 4-year roadmap into 6 months to power AI

“This wasn’t just a modernization effort — it was a reset in how we think about data. We accelerated a 4-year roadmap into 6 months by standardizing on a modern architecture, with Fivetran as a critical foundation for making that possible.”
— Jon McManus, Chief Data and AI Officer, Inova Health

Inova Health is a major healthcare provider serving Northern Virginia and Washington, D.C., supporting more than 4 million patient visits annually across a workforce of 26,000 employees.

Unlike most industries, Inova operates across thousands of vendor systems — spanning clinical, imaging, laboratory, financial, and SaaS platforms — each with its own data model, APIs, and constraints. This level of complexity makes unified analytics at scale inherently difficult.

With thousands of data consumers and growing demand for analytics, Inova needed a better way to unify all of their data across the organization. Information was fragmented across on-prem systems, Epic, SharePoint, APIs, flat files, and vendor platforms, making it difficult to deliver timely, reliable insights. 

As demand increased, delivery timelines stretched to months, and a backlog of hundreds of data requests continued to grow. The organization set out to fundamentally rethink its data foundation, prioritizing a new approach that favored speed, scalability, and long-term flexibility.

Inova’s approach centered on 3 priorities:

• Accelerate cloud migration on a compressed timeline
• Avoid vendor lock-in with a flexible, future-proof architecture
• Enable scalable data replication for modern analytics and AI initiatives

To execute on these, Inova standardized on a modern data architecture with Fivetran, dbt, and Databricks. 

Standardizing on a modern architecture: Fivetran, dbt, Databricks

Fivetran became the foundation for data movement across the organization — replacing fragmented pipelines with a consistent, automated model that could scale across departments and use cases. With Fivetran, the organization unified dozens of data sources in just 6 months — an effort originally expected to take 4 years.

“Fivetran plays a key role in helping us move and manage data at Inova. What used to require custom pipelines, manual effort, and months of work is now standardized, automated, and delivered in near real-time. It fundamentally changed the speed at which the business can operate.”
— Clay Townsend, Data Architect of Enterprise AI, Inova Health

This shift immediately enables use cases that were previously not possible. A key example was Adobe Experience Platform. The initiative had stalled for months due to fragmented data pipelines, complex APIs, and a lack of monitoring across systems. With Fivetran, Inova unified the required data tables and operationalized the platform in under a week — enabling the marketing team to move forward with high-value analytics initiatives.

The organization eliminated one-off integrations entirely, standardizing all data movement through a single, governed ingestion model. To extend this model across the entire organization, Inova leveraged Fivetran’s Connector SDK, enabling the team to integrate custom, proprietary, high-volume data sources without building and maintaining bespoke pipelines.

For systems with complex networking and vendor access constraints, Inova used Fivetran’s Hybrid Deployment to simplify connectivity across highly controlled healthcare environments. This allowed them to centralize business-critical data while reducing operational friction and maintaining strict security and compliance requirements.

Together, these capabilities allowed Inova to extend a single, governed data movement model across both standard SaaS applications and complex internal systems, eliminating fragmentation and reducing long-term operational overhead.

dbt introduced a shared layer for transformation and governance, enabling teams to define consistent metrics and improve trust in data across the enterprise. Databricks provided the scalable platform needed to support advanced analytics and AI workloads.

Driving measurable impact across the enterprise

The impact was immediate and measurable. With Fivetran, Inova eliminated $800,000 in third-party data spend, reduced data movement costs by up to 8x, and cleared a backlog of more than 500 data requests.

Teams moved from months-long delivery cycles to near real-time access to trusted data, accelerating decision-making across clinical operations, finance, and patient care. The organization also dramatically improved delivery speed. A key Emergency Department utilization dashboard was rebuilt in just 2 days, providing near real-time visibility into critical operational metrics.

Today, thousands of users across the organization access self-service dashboards and reports powered by near real-time data — enabling faster, more informed decision-making across clinical operations, finance, and patient care. With timely visibility into metrics like Emergency Department utilization and patient flow, teams can better anticipate demand, allocate resources, and reduce operational bottlenecks. This ultimately allows Inova to respond more quickly and effectively to patient needs while improving efficiency across the system. 

A team of just 6 engineers stood up and now operates the Fivetran platform, scaling from dozens of data sources today to hundreds over time.

This shift has also enabled a new operating model for analytics. Instead of building one-off reports, teams can now create scalable, governed data products that support self-service data access across the organization, reducing duplication and increasing consistency.

Building an AI-ready foundation for intelligence and operations 

This transformation was not just about improving data access. It was about preparing Inova for the next generation of AI-driven applications. By centralizing and standardizing data movement with Fivetran, Inova has created an open foundation that supports AI agents, copilots, and intelligent applications built on trusted, governed data. 

Beyond data ingestion, Inova is extending this foundation into AI development workflows. With support for transformations, MCP-compatible architectures, and downstream activation, Inova can connect data directly to action — powering natural language interfaces, dynamic dashboards, and operational workflows. This is especially critical in healthcare environments, where AI systems must operate on consistent, secure, and governed data.

“Our focus now is on how we operationalize AI across the enterprise. By standardizing and centralizing our data with Fivetran, we’re creating the foundation for AI agents and applications that can act on trusted, governed data — not just generate insights, but drive action.”
— Jon McManus, Chief Data and AI Officer, Inova Health

With a modern data foundation in place, Inova is positioned to scale intelligence across the enterprise — delivering faster insights, improving operations, and enabling the next wave of AI innovation.