Revenue Leakage Detection
National Freight & Logistics Provider — top 20 US 3PL, $2.8B annual freight spend.
2025
The challenge
What we found.
A national logistics company managing $2.8B in annual freight spend discovered they had no systematic way to verify they were being billed correctly. Their contracts contained complex terms including volume-based tier discounts, fuel surcharge caps, and accessorial charge limits.
Manual audits caught obvious errors but missed sophisticated discrepancies. Industry benchmarks suggested they were likely losing 2–4% of freight spend ($56M–$112M) to billing errors and missed contract terms.
The approach
How we built it.
PRR built an AI-powered Contract Compliance & Revenue Recovery system:
- Used NLP to extract terms from 340+ contracts into structured, queryable data formats
- Real-time comparison of every invoice line item against applicable contract terms
- ML models identifying unusual patterns suggesting billing errors (anomaly detection)
- Automated dispute generation workflow with supporting documentation for identified errors
- 01DigitizeContract NLP extraction
- 02MatchInvoice vs. contract terms
- 03RecoverAutomated dispute filing
The outcome
The impact.
| Metric | Before | After | Improvement |
|---|---|---|---|
| Invoice audit coverage | 12% (sampling) | 100% | Full coverage |
| Identified discrepancies | $3.2M/year | $47M/year | 14.7× increase |
| Successfully recovered | $0.8M/year | $41M/year | +$40.2M |
| Time to identify issues | 45–60 days | Real-time | Immediate |
Stack
Tools used.
Keep reading
Related case studies.
Public Sector · Human Services
Statewide crisis response network
Cut a nine-hour survivor vetting cycle to seconds across 57 counties on a zero-trust edge.
Read the case studyFinancial Services
Financial forecasting engine
Real-time time-series forecasting with ERP integration that surfaced $25M in annual savings.
Read the case studyRetail
Inventory optimization system
Predictive analytics and WMS integration that turned inventory 18% faster across the chain.
Read the case study