Medicare Provider Utilization & Access Dashboard
Transforming Public Healthcare Data into Executive Decision Support
Healthcare organizations collect enormous amounts of data, but raw information alone rarely helps leaders make better decisions. This project explores how public CMS provider utilization data can be transformed into an interactive executive dashboard that helps users identify meaningful trends, understand provider variation, and investigate opportunities for deeper analysis.
Rather than building a dashboard around charts alone, the project begins with a more important question:
What decisions should this dashboard help someone make?
Product Design
The dashboard is organized around the questions an executive or product leader is likely to ask rather than around the structure of the underlying dataset.
The experience begins with an executive overview highlighting utilization trends, provider distribution, payment variation, and specialty-level metrics. Users can then progressively drill into geographic regions, provider specialties, procedures, and individual providers to investigate specific patterns.
Rather than overwhelming users with every available metric, the dashboard emphasizes a small set of carefully defined measures supported by clear definitions, filtering, and visual exploration.
The design intentionally distinguishes between:
Outcome Metrics — What is happening?
Diagnostic Metrics — Why might it be happening?
Data Quality & Context — How much confidence should we have in the information?
This approach reflects my belief that analytics products should reduce cognitive burden by helping users recognize meaningful patterns before asking them to interpret raw data.
This dashboard was intentionally designed around executive questions rather than database tables.
Instead of asking:
"What fields are available?"
the design begins with questions such as:
Where are utilization patterns changing?
Which provider specialties warrant further investigation?
Where does geographic variation appear significant?
Which metrics should leadership monitor over time?
Only after those questions are defined does the dashboard determine which data should be surfaced to support them.
Technologies Used
CMS Open Data
Google BigQuery
Google Cloud Storage
Looker Studio
SQL
Data Modeling
Executive Dashboard Design
Metric Definition
Healthcare Analytics
Data Governance
Product Analytics
The Problem
Healthcare analytics often fail not because organizations lack data, but because decision-makers struggle to identify which information matters most.
Large public datasets contain thousands of measures across providers, procedures, payments, specialties, and geographic regions. Without thoughtful organization, users are forced to search through reports instead of focusing on the questions they are trying to answer.
This project explores how product thinking can transform complex public healthcare data into an experience that supports investigation rather than simply displaying information.
Product Goals
Transform large public CMS datasets into meaningful executive insights.
Highlight provider utilization, payment, and geographic patterns.
Support progressive drill-down from high-level trends into detailed investigation.
Present information using trusted definitions and consistent metrics.
Help users identify opportunities for additional analysis rather than drawing unsupported conclusions.
Demonstrate how analytics products should support decision-making instead of simply reporting data.
What This Project Reinforced
Working with large public healthcare datasets reinforced another principle that has guided much of my career: data only becomes valuable when it helps someone make a better decision.
The technical work involved importing, modeling, and visualizing CMS data, but the more meaningful product work involved deciding which measures mattered, how they should be defined, how they related to one another, and how users should navigate from a broad trend to a specific question requiring further investigation.
Throughout my career, I have found that successful analytics products are not measured by the number of reports they produce or the number of charts they display. They are measured by how effectively they help people recognize what matters, understand the context surrounding it, and determine the next question they should ask.
That philosophy extends beyond dashboards. Whether designing analytics, AI assistants, or enterprise healthcare products, I believe technology creates the greatest value when it reduces complexity, builds trust in information, and helps people apply their expertise where it matters most.