In Moore v Publicis Groupe & MSL Group, U.S. Magistrate Judge Andrew Peck (U.S. Magistrate Judge, Southern District of New York) recognized “that computer-assisted review is an acceptable way to search for relevant ESI in appropriate cases,” which could have important ramifications for the punishingly expensive process of document review in all large U.S. civil litigation.
Computer-assisted review, also known as predictive coding, uses algorithms to determine relevance, privilege, etc. This is generally how it works: A small set of documents is reviewed and coded by high level reviewers. The program analyzes the documents and then predicts the reviewer’s coding on the next small set of documents. This is done for multiple iterations; in Moore, they agreed to do this seven times. When the results sufficiently match (they agreed to a 95% confidence level in the Moore case), the program then codes the entire review set.
As Mag. Judge Peck’s opinion points out, all of us who use a spam filter use a form of predictive coding.
According to the Sedona Proclamation Model (the Sedona Conference is the cutting edge of U.S. e-discovery practice and scholarship), lawyers should advise opposing counsel of a plan to use predictive coding and seek agreement.
The Court reminded the parties that computer-assisted review “works better than most of the alternatives, if not all of the [present] alternatives. So the idea is not to make this perfect, it’s not going to be perfect. The idea is to make it significantly better than the alternatives without nearly as much cost.”
The Court rejected the Plaintiffs’ argument that the use of predictive coding did not comply with Federal Rule of Civil Procedure 26(g)(1)(A), which requires that an attorney certifies that a production is complete and correct. The Court also found that Plaintiffs’ 702 and Daubert argument did not apply to how documents are searched and identified in discovery.
Rule 1, which is to “secure the just, speedy, and inexpensive determination” of lawsuits. Fed. R. Civ. P. 1. That goal is reinforced by the proportionality doctrine set forth in Rule 26(b)(2)(C).
Mag. Judge Peck determined that the use of predictive coding was appropriate considering:
(1) the parties’ agreement,
(2) the vast amount of ESI to be reviewed (over three million documents),
(3) the superiority of computer-assisted review to the available alternatives (i.e., linear manual review or keyword searches),
(4) the need for cost effectiveness and proportionality under Rule 26(b)(2)(C), and
(5) the transparent process proposed by the defendants.
Some believe that predictive coding is a step in the right direction for a great deal of significant U.S. civil litigation because large litigation should be decided on the merits rather than based on the punishing and extremely high cost of discovery inherent in large document productions. In traditional large-scale litigation, litigants using standard document review and production processes require attorney review of millions of pages (or more) of documents. As a consequence, the costs of discovery, regardless of the merits, may drive the litigation.