Case study · Healthcare provider
Revenue Cycle Optimization
A machine-learning model found 500+ treatments that were delivered but never billed — and helped file them.
Client
A community hospital in a large healthcare system
Industry
Healthcare provider
Result
20% improvement in operating margin
Capability
Analytic & Consulting Services
The situation
What the client was facing.
The hospital suspected that legitimate treatments had been delivered without claims being submitted to insurers. It wanted to find those unbilled charges and recover the revenue.
Missed charges hidden across departments
Thousands of diagnosis and charge codes
Every claim had to be legitimate
Our approach
How we got there.
Methods used
- 01
Built a machine-learning association model linking diagnostic codes to CPT charge codes
- 02
Applied it across departments including day surgery, rehab and cardiology
- 03
Flagged cases with a strong association but no charge — e.g. a cataract operation with no lens charge
- 04
Sized missing charges by specialty
- 05
Produced diagnostics and recommendations to prevent future misses
Value created
20%
improvement in operating margin
More than 500 cases found where treatment was given but not charged
Missing charges segmented by specialty, with root causes
Each missing charge validated before filing
Filing them increased operating margin by 20%
More client work
Related case studies.
$100MM
improvement in marketing ROI
Marketing Optimization
Modeled what drives inquiries, pre-qualified buyers and sales — nationally and by DMA — and rebalanced marketing spend.
Read the case →$50–70MM
profit improvement
Customer Segmentation and Profitability
Voice-of-the-customer research and customer-level profitability, turned into seven segments and a plan to win back share.
Read the case →$75MM
revenue growth
Pricing Strategy
Replaced deal-by-deal, one-size-fits-all pricing with a tiered, elasticity-based strategy across the portfolio.
Read the case →
Want a result like this one?
A 30-minute walkthrough of Gazelle on a category like yours.