Case Study: Foley & Lardner achieves $10M savings and 10× faster investigation with Relativity's Active Learning (RelativityOne)

Foley & Lardner Uses Active Learning to Save $10 Million on a White-Collar Investigation

Foley & Lardner faced a government white‑collar investigation that ballooned from 8 million to 117 million documents after a single text message suggested anticompetitive behavior. The firm needed to review the data thoroughly and quickly without missing key materials, so they turned to Relativity and its RelativityOne platform with active learning to tackle the massive review task.

Relativity implemented a novel active‑learning workflow—splitting the corpus into 10 million‑document sets with multiple analytics indexes, running parallel active‑learning projects, and using the prioritized review queue—to surface the most likely responsive material. That approach identified about 3 million likely responsive records, bubbled 23,000 documents for immediate review and production, found thousands of responsive docs missed by search terms, and helped Foley & Lardner complete the investigation ten times faster while saving an estimated $10 million, with Relativity’s platform and support cited as critical to the outcome.

Foley & Lardner

Nicholas Cole

Director of Litigation Support