The February 24 decision by the Southern District of New York in a case involving the use of computer-assisted review in electronic document productions may well pave the way to broader acceptance of this technology in e-discovery. In his ruling in Da Silva Moore v. Publicis Groupe and MSL Group, Magistrate Judge Andrew Peck recognized that computer-assisted review is an acceptable way to search for relevant electronically stored information (ESI) in appropriate cases.
Computer-assisted search and coding, also known as "predictive coding," deploys software tools that use complex algorithms to determine the relevance of a large collection of documents based on a reviewer's coding of a small sample of such documents. Thus, an attorney with knowledge of the facts of a case need only examine a relatively small sample of documents for relevance; once the sample is reviewed and coded, the program analyzes the documents and then predicts the reviewer's coding on the next small set of documents. After multiple iterations, when the results sufficiently match, the program codes the entire review set. The process can potentially save hundreds or thousands of hours of human review time—and accompanying expenses—that otherwise would be wasted on insignificant materials.
In Da Silva Moore, a gender discrimination case, both parties had agreed to use computer-assisted review but disagreed over the appropriate methodology. The defendants proposed using a number of rounds to test and refine the searches to stabilize the review software. They also planned to share the seed documents and the documents that the software tagged relevant and irrelevant with the plaintiffs, and then allow the plaintiffs to offer feedback as they refined the process. The plaintiffs claimed the defendants' proposed protocol was not as reliable or transparent as it should be. At a hearing on the matter, the judge asked the parties to submit a joint protocol governing the use of computer-assisted review, which became part of his order.
Judge Peck approved the defendant's use of computer-assisted review in this case for five reasons: (1) the parties' general agreement to use computer-assisted review, (2) the large volume of ESI involved in the case, (3) the superiority of computer-assisted review in this case to the available alternatives, such as manual review and keyword searches, (4) the need for cost effectiveness and proportionality under the Federal Rules of Civil Procedure, and (5) the transparency of the proposed computer-assisted review process. The judge noted, however, that his opinion "does not mean computer-assisted review must be used in all cases, or that the exact ESI protocol approved here will be appropriate in all future cases that utilize computer-assisted review."
"What the Bar should take away from this Opinion," the judge wrote, "is that computer-assisted review is an available tool and should be seriously considered for use in large-data-volume cases where it may save the producing party (or both parties) significant amounts of legal fees in document review."
Parties to litigation in the U.S. have been reluctant to embrace new e-discovery technologies to assist in identifying relevant documents, mainly because of concerns that the software might miss documents that are critical to the case. Now that a leading federal court has endorsed the evidence that computer-assisted review may in some cases be more accurate than the traditional document-by-document review, the tide may begin to turn, ushering greater acceptance of the new technology—not least because of its efficiency and the remarkable cost-reductions it makes possible.
Faegre Baker Daniels has been exploring the use of computer-assisted review in appropriate circumstances. The technologies at our disposal are Relativity and Axcelerate, the latter of which was under consideration in DaSilva. For more information about our computer-assisted review tools, contact Client Technology Solutions.