It's no secret that discovery in litigation can be extremely — sometimes prohibitively — expensive. A recent study of electronic discovery costs shows that the total cost of production could range from $17,000 to $27 million, with a median of $1.8 million.1 Because discovery expenses have the potential to explode, it often becomes the predominant focus and concern in litigation, demanding the lion's share of the party's resources. Recent advancements in e-discovery technology may provide a way to manage the costs, thus shifting the party's resources and focus back to the legal issues in the litigation.

One such advancement is the use of predictive coding. Predictive coding is a computer-assisted review method that substantially reduces, but does not entirely eliminate, the need for human linear review. At its most basic, predictive coding uses computer algorithms and logic to analyze and mine electronically stored data so that responsive documents can be identified among the sea of stored electronic data. An attorney familiar with the issues in the case reviews a small subset of the stored data, and a mathematical model of relevant documents is generated from this direct input. This model is used to identify the relevant documents, then segregate them from those that are irrelevant in the main collection. Studies have shown that predictive coding is at once faster and more effective than what has to-date been the gold standard in discovery — human review and keyword searching. This translates to significant cost savings for the client because less attorney time is needed to review a smaller, more targeted set of documents.

Use of predictive coding is gaining acceptance in federal and state courts. On February 24, 2012, Magistrate Judge Andrew Peck of the Southern District of New York issued the first federal opinion and order in the nation to approve predictive coding as an acceptable method of searching for responsive documents in Da Silva Moore v. Publicis Groupe & MSL Group, 11 Civ. 1279 (ALC) (S.D.N.Y. Feb. 24, 2012). Judge Peck's decision has been subsequently affirmed by District Court Judge Andrew Carter.

Most recently, in Global Aerospace, Inc. v. Landow Aviation, LP, Consolidated Case No. CL 61040, in the Circuit Court of Loudoun County, Virginia, a Schnader team led by Thomas C. Gricks III (Pittsburgh) and Jonathan M. Stern (Washington, D.C.) won the first state court approval to use predictive coding for processing and production of electronically stored information absent agreement with the opposing party. Our client had about 250 gigabytes of reviewable electronically stored information, which could easily translate to about two million documents and require 20,000 hours of human linear review. Schnader's team successfully demonstrated to the court that predictive coding would reduce the time to cull the relevant documents to two weeks and at 1/100th of the cost.

Therefore, a protocol based on predictive coding should be used instead of human linear review or iterative keyword searches to avoid unnecessarily burdensome and costly discovery. More importantly, the team successfully persuaded the Court that the use of predictive coding met Virginia's reasonableness standard for discovery.

While the benefits of predictive coding are most dramatic in cases involving large amounts of reviewable electronic data, the savings in time and cost are universal. Contact Schnader's e-Discovery practice group to see how predictive coding can help rein in the discovery costs in your case.

Footnote

1. Nicholas M. Pace & Laura Zakaras, Where the Money Goes: Understanding Litigant Expenditures For Producing Electronic Discovery, 17 (2012)

www.schnader.com

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.