How Not to Thwart the Progress of Technology-Assisted Review
Recent case law has shown strong support for the use of technology-assisted review (TAR): its accuracy and efficiency have been praised by judges and parties alike. However, despite her approval of the process as a “far more accurate means of producing responsive ESI in discovery … than human review or keyword searches,” Magistrate Judge Peggy A. Leen rejected the plaintiff’s use of the tool in Progressive Casualty Insurance v. Delaney after many months of delay in the production of requested electronic evidence.
Before commencing discovery, the parties agreed to a protocol for handling electronically stored information (ESI), which the court entered as an order. In the order, the parties had negotiated and agreed to search terms to apply to the collection. Subsequently, Progressive collected 1.8 million documents and applied a keyword search that resulted in 565,000 potentially responsive documents. Progressive had the option to produce the documents pursuant to a clawback or to review them manually. The company elected manual review, and it reviewed 125,000 documents before unilaterally deciding that it would be too time-consuming and expensive to continue.
Despite the agreed-upon order, Progressive decided to employ TAR to find responsive documents in the pool of 565,000 “hit” documents, not the entire corpus. The FDIC-R balked and asked the court to order Progressive to produce the 565,000 documents subject to a clawback for privileged documents, which would eliminate the expense of manual review for Progressive, speed discovery, and eliminate the disputes over TAR. Alternatively, it suggested an alternative TAR protocol that was more transparent and collaborative. Ultimately, the court ordered Progressive to produce all 565,000 documents—and to withhold those documents identified by privilege filters—without further review.
Parties cannot haphazardly agree to a discovery plan without considering the potential cost and timeline of the options they choose. Here, Progressive made a number of missteps that sealed its fate: it refused “to engage in the type of cooperation and transparency that its own e-discovery consultant has so comprehensibly and persuasively explained is needed.” Specifically, it retained a discovery expert only after agreeing to the protocol that became a court order. Working with an e-discovery consultant to run the search terms against the collection before agreeing to a protocol could have led to greater foresight about the cost and scope of discovery and ultimately to significant savings; a seasoned e-discovery specialist could also have constructed a defensible, transparent strategy for employing TAR. Furthermore, Progressive failed to obtain the plaintiff’s or court’s agreement to amend the court order to permit the use of TAR instead of manual review, did not involve the FDIC-R in any decisions regarding its TAR methodology, and deviated from the vendor’s best practices for using TAR. Parties must realize that cooperation does not end once a discovery protocol is agreed upon; maintaining proactive communication throughout the discovery process can improve the chances of reaching a solution that will appease both sides.
Laurie Stoni is an eDiscovery Consultant with Conduent. She can be reached at email@example.com.