A recent decision of the United States District Court, Northern District of California affirmed that a litigant has no right to compel the other party to use technology assisted review (“TAR”) or predictive coding* to identify relevant documents. (See also Hyles v. New York City, 2016 WL 4077114, at *2-3 (S.D.N.Y. Aug. 1, 2016); and In re Biomet M2a Magnum Hip Implant Prod. Liab. Litig., 2013 WL 1729682, at *2-3 (N.D. Ind. Apr. 18, 2013)).
In seeking an order compelling the Defendant Pfizer, Inc. (“Pfizer”) to use predictive coding/TAR, the Plaintiff argued that predictive coding/TAR is a more sophisticated tool than the traditional application of keyword search terms and sought to have representatives of both parties involved in the process of developing the search process. The Plaintiff said this would save time and money for both sides.
Pfizer proposed an alternative method, which involved the iterative application of search terms and “rigorous sampling” to verify results. Pfizer agreed to share proposed search terms with the Plaintiff in advance, to apply search terms agreed upon by both sides and to consider the inclusion of any other search terms proposed by the Plaintiff.
In refusing to compel Pfizer to participate in the Plaintiff’s TAR-based process, the Court noted that there is no precedent for ordering a party to engage in technology assisted review and that the party itself is best situated to decide how to search for and produce information responsive to discovery requests.
As the court reasoned in Hyles, the responding party is the one best situated to decide how to search for and produce electronically stored information responsive to discovery requests. As such, the responding party (citations omitted):
can use the search method of its choice. If [the propounding party] later demonstrates deficiencies in the . . . production, the [responding party] may have to re-do its search. But that is not a basis for Court intervention at this stage of the case.” Id. “[I]t is not up to the Court, or the requesting party . . ., to force the . . . responding party to use TAR when it prefers to use keyword searching. While [the propounding party] may well be correct that production using keywords may not be as complete as it would be if TAR were used . . ., the standard is not perfection, or using the “best” tool . . ., but whether the search results are reasonable and proportional.
Although predictive coding and other forms of technology assisted review are firmly established in the jurisprudence as acceptable processes of identifying responsive electronically stored information (ESI) (See Da Silva Moore v. Publicis Groupe, 2012 U.S. Dist. LEXIS 23350 (SDNY, Feb. 24, 2012)), the Court noted the absence of any basis on which it could make the order sought in absence of any evidence that Pfizer’s preferred method would produce or has produced insufficient discovery responses.
In short, litigants seeking a cooperative predictive coding/TA-based discovery process will need the agreement of the other party – at least in the US. Alternatively, organizations must be prepared to present evidence as to why the other party’s preferred process will produce inadequate results. Expert evidence relating to the nature of the document set and fallibility of keyword search terms in the specific circumstances may be useful in this regard.
*TAR, or predictive coding, is the use of keyword search, filtering and sampling to automate (usually via machine learning) portions of an e-discovery document review. The goal of predictive coding is to reduce the number of irrelevant and non-responsive documents that need to be reviewed manually.