How Vizient Reduced Critical Analytics Workloads with Starburst

About

Vizient is the nation’s largest provider-driven healthcare performance improvement company, helping healthcare organizations improve quality, operational performance and cost efficiency through analytics, supply chain solutions and advisory services. With a history of growth through mergers and acquisitions, Vizient’s large and heterogeneous data estate is continually expanding and the internal data marketplace team is responsible for building reusable patterns and data assets that enable business units across the organization to deliver outcomes without each domain having to rebuild the same capabilities from scratch.

Challenge

After several M&As, Vizient had a polyglot data ecosystem, on-premises sources, cloud systems, and a wide variety of inherited database technologies, including relational datastores and more. Getting data available within a single domain was complex enough; combining it across domains for more sophisticated analytics meant bespoke ETL integrations, significant data movement, and long waits on engineering teams before any insight could be uncovered.

“Cross-domain analytics at scale — and being able to govern it — was the challenge that brought us to look at Starburst.” — Ram Radhakrishnan, Engineering Leader, Tech Core & Platforms, Vizient

Solution

Vizient ran an opinionated POC to validate Starburst for federation across on-premises and cloud sources, as well as SQL compute on their Azure data lake. The results proved the strategy. Starburst became the SQL backbone of the Enterprise Data Library (EDL), Vizient’s internal data marketplace, enabling cross-domain analytics.

On top of that foundation, Vizient built a self-service marketplace of data producers and consumers, with governed sandboxes for rapid prototyping and a structured path to publish trusted data products. Starburst’s built-in RBAC handles access control across the stack, and its deployment flexibility — self-hosted in Vizient’s own Virtual Private Cloud (VPC) with a path to fully managed — was a key factor in a regulated environment.

Results

The impact was most visible in Vizient’s ambulatory analytics team, which produces periodic benchmarking reports across tens to hundreds of hospital systems. The legacy process required hours of time-series number crunching across their original SQL environment.

With Starburst, the same workflow, with only minor dialect adjustments for Trino, now completes in under five minutes.

“They were checking their code to make sure they hadn’t introduced a new bug because they couldn’t believe it had gone from 12 hours to less than five minutes. It was just the sheer reduction in time.” — Ram Radhakrishnan, Engineering Leader, Tech Core & Platforms

Broader outcomes across the organization include:

• 12 hours → under 5 minutes for hospital system benchmarking workloads, enabling faster delivery of performance insights to healthcare organizations
• Self-service data discovery in minutes — any team can find and access trusted data assets across the enterprise without waiting on engineering
• Eliminated bespoke ETL for cross-domain analytics — federated queries replace data movement for ad hoc and recurring analytical workloads

As Vizient continues to build toward conversational analytics and AI-embedded workflows, the EDL’s data product layer, designed to surface business context and metadata for human users, is well positioned to serve as the foundation for AI-ready data access.

More resources: How Businesses Are Laying a Trusted Data Foundation, Datanova 2026