In this episode, Tom Fox welcomes back Vince Walden, CEO of konaAI, which is the sponsor of this podcast series. Vince is well known for his leadership in data analytics, machine learning, and AI, and we take a deep dive into all of these topics through the lens of the Scoular FCPA Enforcement action.
The Scoular case is a unique learning tool for using data analytics, machine learning, and AI for compliance. In this matter, roughly 2,000 near-$2,000 payments were tied to customs brokerage activity. This case illustrates broader uses of data analytics beyond numbers, including mining unstructured text in invoice and payment-description fields (e.g., repeated terms like “re-inspection fee” and Spanish phrases) and linking it to structured AP data.
Walden explains how combining structured and unstructured data reduces investigator bias, how round-dollar and repetitive payments to high-risk vendors can be continuously risk scored using hundreds of tests, and how machine learning can “find more like this” across large transaction populations. We discuss integrating communications data when available, using monitoring within typical 30–60-day payment cycles to prevent payments, supporting self-disclosure decisions amid DOJ guidance, and launching a minimum viable analytics program by starting with AP spend, invoices, POs, and payments pulled from ERP systems.
Key highlights:
- Structured vs. Unstructured Data
- Red Flags Round Dollars
- Text Mining and Three Lines
- Transactional Fingerprints
- Machine Learning for Compliance
- Linking Comms and Payments
- Prevention Through Monitoring
- Continuous Improvement and ROI
- Self-Disclosure and Culture Data
- 90-Day Analytics Roadmap
Resources:
Vince Walden on LinkedIn
Tom Fox
To learn about the intersection of Sherlock Holmes and the modern compliance professional, check out Tom’s latest book, The Game is Afoot-What Sherlock Holmes Teaches About Risk, Ethics and Investigations on Amazon.com.
