Discover how Cost Intelligence accelerates hospital cost reduction and modern healthcare cost management by turning activity-based cost data into prioritized, defensible action.
Download
Health systems don’t struggle to access cost data.
They struggle to trust it, explain it, and act on it.
For years, hospitals have relied on costing solutions that average away reality. A surgical supply that costs $2 in one case and $8 in another becomes “$6.” True cost variation analysis becomes impossible. Clinicians disengage. Finance teams issue blunt mandates. Improvement stalls, not because people don’t care, but because the data lacks the credibility required to drive change.
Health Catalyst built PowerCosting to solve the first and most important problem: establishing true, activity‑based cost at the level clinicians recognize as real.
But over time, we learned something important. Accurate cost is necessary. Health systems didn’t just need cost visibility. They needed a way to translate cost truth into prioritized defensible action without months of manual analysis or armies of analysts.
That realization led us to build Cost Intelligence, and it is why we made the deliberate decision to rebuild the foundation of PowerCosting, and power Cost Intelligence, on Databricks.
This was not a technology refresh. It was a redefinition in how improvement is delivered.
Cost Intelligence is not another costing system.
It is an intelligence layer built on top of PowerCosting that continuously analyzes activity‑based cost data to surface where variation is real, material, and worth acting on—especially for service line leaders and clinicians.
PowerCosting establishes the truth:
Cost Intelligence turns that truth into prioritized action by:
In short: PowerCosting shows what care truly costs. Cost Intelligence shows where to act first—and why.
Delivering on that promise required a fundamentally different approach to analysis.
Historically, identifying meaningful cost improvement opportunities has been slow and human‑intensive.
Health systems assembled expert teams—finance analysts, clinicians, and often external consultants—to answer a simple but critical question: Where should we focus first? The work that followed took months. Data was extracted, reconciled, debated, sliced, re‑sliced, and pressure‑tested in nonstop meetings. Even then, results were often challenged by clinicians who questioned how they were derived.
The constraint wasn’t expertise. It was scalability.
Cost Intelligence was designed to change that dynamic entirely.
At the core of Cost Intelligence is a set of proprietary, trained improvement models, built from more than $2.2 billion in proven healthcare improvement work delivered across hundreds of health systems. These models capture what experts have learned in the field: where variation truly matters, which patterns lead to sustainable savings, which trade‑offs to avoid, and how to distinguish signal from noise.
In the past, applying that knowledge required humans to interrogate data manually. Today, those same patterns are embedded directly into the system.
Because Cost Intelligence runs on Databricks, it applies these trained improvement models directly against full‑fidelity, encounter‑level cost data, scanning millions of records in seconds. What once required weeks or months of expert effort now happens continuously and automatically:
This is the critical shift: from improvement as a project to improvement as a capability.
Cost Intelligence does not replace human expertise—it amplifies it. Experts no longer spend months hunting for signal. They begin with a short, defensible list of high‑impact opportunities and focus on what actually drives results: validating insights, engaging clinicians, and executing change.
Speed matters here not just because it is efficient, but because it changes behavior. When insights arrive in seconds instead of months, momentum builds. Conversations shift from “Do we trust the data?” to “What do we do next?”
Improvement becomes operational, not theoretical.
Delivering improvement at this speed requires more than strong models. It requires an architecture capable of applying those models repeatedly to real, granular data—at scale.
Databricks provides that foundation.
PowerCosting calculations now run natively on Databricks, persisting outputs as Delta Lake tables. Cost data remains granular, traceable, and directly joinable to shared clinical dimensions such as procedure, provider, and encounter. It is not summarized into static reports or exported into disconnected systems.
Cost Intelligence operates on those same datasets. Instead of analyzing pre‑aggregated reports, it continuously evaluates the full population of costed activity as new data arrives and models evolve. Databricks combines elastic compute with open, high‑performance data storage to enable this continuous, population‑scale analysis.
Just as importantly, Databricks allows Cost Intelligence’s improvement models and AI‑driven logic to run where the data already lives. Feature engineering, scoring, and prioritization happen in the same environment as the underlying cost truth. There is no fragile data movement between platforms, no manual hand-offs, and no re‑engineering required as new use cases emerge.
The result is a tight, scalable loop between data, intelligence, and action.
Rebuilding on Databricks did more than modernize PowerCosting. It changed what it could become.
PowerCosting is no longer a static costing engine. It is:
Cost Intelligence is the first—and most visible—expression of that transformation.
Cost data is no longer the end state. It is the starting point of an ongoing improvement cycle:
That cycle only works if the platform can keep pace. Databricks makes that possible.
For health systems, this changes the economics of improvement. The impact is tangible.
They move:
Most importantly, they gain confidence—not just in the numbers, but in the decisions those numbers support.
In a market saturated with analytics and AI claims, confidence comes from defensibility. Insight must be grounded in reconciled cost, real clinical activity, and proven evidence of what works.
That is what Cost Intelligence—powered by Databricks—delivers.
At HIMSS, I’ll share how this evolution is playing out in practice—and what it looks like when improvement becomes a system capability, not a consulting project.
In the Databricks booth theater session, we’ll explore:
If you lead analytics strategy, cost transformation, or enterprise data platforms, this conversation is built for you.
Healthcare does not need more dashboards. It needs defensible insight delivered at the speed of decision-making.
We hope you’ll join us at the Databricks HIMSS Booth #5453, Wednesday, March 11 from 3:00 – 3:30 PM.
