Case Study Deskcasestudydesk.com
Legal (law firm)Sourced

Case Study: Dinsmore & Shohl reaches 98.2% recall on 26,000-contract cyber review with Everlaw

Dinsmore & Shohl Case StudySourced & dated by Case Study Desk
Key facts · TL;DR
Company
Dinsmore & Shohl
Industry
Legal (law firm)
Challenge
Review 26,000 contracts to find every clause requiring security-incident notice, where incomplete notice creates legal exposure.
Headline result
Dinsmore & Shohl combined EverlawAI Coding Suggestions and Predictive Coding to hit 98.2% recall reviewing 26,000 contracts

Key results

98.2%
Final recall rate
after Predictive Coding
~20,000
Contracts eliminated up front
of 26,000
~260 hrs
Attorney hours saved
over 6 weeks of full-time work

The challenge

In a cyber incident response, Dinsmore & Shohl had to rapidly review 26,000 contracts to identify every clause requiring notification of a security incident. Contractual notice obligations were absolute, so incomplete notification would create additional legal exposure.

The solution

The firm used a layered, three-phase approach: strategic search to eliminate drafts and duplicates, EverlawAI Coding Suggestions for pattern recognition, and Predictive Coding as a final validation layer, with human-in-the-loop review before each phase advanced.

When you're talking about notifying counterparties to whom you have a contractual obligation... 80% doesn't really do it.

PP
Peter Pepiton
Ediscovery Director, Dinsmore & Shohl

The results, in context

The initial search phase eliminated roughly 20,000 of the 26,000 contracts. Coding Suggestions reached about 92-93% recall, and Predictive Coding lifted the final recall rate to 98.2%. The team estimated manual review would have taken about 260 attorney hours, more than six weeks of full-time work.

Products used

Everlaw EverlawAI AssistantEverlaw Coding SuggestionsEverlaw Predictive Coding