While a party to litigation may be interested in using predictive coding to drive down the costs of reviewing electronically stored information (“ESI”), the question often arises as to whether the use of predictive coding is defensible in a court of law. In other words, if a party decides to use predictive coding, can the use of the technology be challenged and even thwarted?

This month’s column will address the defensibility of predictive coding in court cases, whether the use of predictive coding should be disclosed to the opposing party, and how to address challenges to the use of predictive coding.

Cases Endorsing the Use of Predictive Coding

There are recent court decisions endorsing the use of predictive coding. In Monique v. Da Silva Moore, the U.S. District Court for the Southern District of New York ultimately endorsed the use of predictive coding. This is the first opinion to date to do so.

In Global Aerospace v. Landow Aviation Limited Partnership, the Circuit Court in Loudoun County, Virginia endorsed the use of predictive coding over the objection of one of the parties. These decisions represent an early acceptance by courts of the use of predictive coding in ESI intensive cases.

Disclosing the Use of Predictive Coding

FREE Investigation Report Template

Prepare thorough, consistent investigation reports with our free report template.

Download Template

While there are recent court decisions endorsing the use of predictive coding, it is possible that you could find yourself in a situation where the court or opposing counsel is unfamiliar with predictive coding and the circumstances in which it should be deployed in the case. This raises the question as to whether litigants should even disclose the use of predictive coding to the other side.

On the one hand, not disclosing the use of predictive coding can prevent extensive motions practice regarding the use of the technology. According to this school of thought, a party to a case should be able to decide the platform that it wants to use to review ESI and should not need to secure an agreement to use that platform.

On the other hand, some litigants may wish to secure the agreement of the opposing party not only as to the use of predictive coding but also regarding the protocol in the case. This second school of thought suggests that full disclosure, both as to the use of the technology and the protocol, is important in order to avoid surprises at a later point in the litigation. Whether to disclose the desire to use predictive coding is a tactical decision that will have to be made in the overall context of the case.

Handling Predictive Coding Objections

If a tactical decision has been made to use predictive coding and to disclose its use to the opposing party, the opposing party may object to the use of predictive coding. In such a situation, it would be advisable to file a motion for protective order and to secure the court’s agreement as to the use of predictive coding in the case. This will require educating the court regarding the pitfalls of using other search and review methods and the cost considerations for each method.

For example, if the opposing party insists on the use of keyword search, consider highlighting to the court the fact that keyword searches only capture 20 percent of relevant evidence at best. If pure human review of all documents is sought, it is important to point out the extremely high costs associated with such a review and the fact that human review only locates 60 percent of relevant documents to the case.

Predictive coding promises the lowest cost per page to review compared to all other search methodologies and locates between 75-90 percent of relevant documents. By pointing out the fact that predictive coding is more likely to yield relevant documents at a substantially reduced cost, a party can succeed in convincing the court that predictive coding should be used to locate and review ESI in a case.