Last week in Dynamo Holdings et al. v. Comm’r of Internal Revenue, 143 T.C. No. 9, the U.S. Tax Court became the latest court to approve the use of predictive coding in litigation. The scenario addressed by Judge Buch has become familiar: Faced with the review of a large volume of documents for potential production, the producing party (Dynamo) sought to use predictive coding to identify responsive documents; the requesting party (the IRS) opposed the approach, contending that predictive coding is an “unproven technology.” The court rejected the IRS’s position, observing that “although predictive coding is a relatively new technique,” the technology has become “widely accepted for limiting e-discovery to relevant documents and effecting discovery of ESI without an undue burden.”
Judge Buch first considered the role that courts should play in the decision to use predictive coding, and expressed a decidedly hands-off view:
[A]lthough it is a proper role of the Court to supervise the discovery process and intervene when it is abused by the parties, the Court is not normally in the business of dictating to parties the process that they should use when responding to discovery. If our focus were on paper discovery, we would not (for example) be dictating to a party the manner in which it should review documents for responsiveness or privilege, such as whether that review should be done by a paralegal, a junior attorney, or a senior attorney. Yet that is, in essence, what the parties are asking the Court to consider—whether document review should be done by humans or with the assistance of computers.
Despite this viewpoint, the Judge went on to consider the request for permission to use predictive coding, because the subject of computer-assisted review tools had not yet been addressed by that court.
Dynamo presented uncontroverted evidence that predictive coding would reduce the cost of the document review and production in that case by more than 80%. It also showed that predictive coding can eliminate inefficiencies and human error inherent in keyword searching, and offers the ability to use “statistics and other sampling information” to test and verify the production. After hearing this evidence, the court approved the request to use predictive coding. As for the IRS’s concerns about the completeness of the production, the court offered this:
Respondent fears an incomplete response to his discovery. If respondent believes that the ultimate discovery response is incomplete and can support that belief, he can file another motion to compel at that time.
At DiscoverReady, our team of experienced consultants works with our clients to analyze the benefits and drawbacks of using predictive coding for a particular matter. These clients often express the concern that, if they choose the predictive coding route, they will get bogged down in motion practice and endless meet-and-confer sessions with the opposing party and its e-discovery experts. To be sure, in many cases the prudent approach is to seek agreement from the other side on the protocol for using predictive coding, and get buy-in from the court on that agreement – but that approach can indeed cost additional time and money.
But for clients with the right risk tolerance and the appropriate litigation strategy, we suggest they adopt Nike’s mantra, impliedly endorsed by the Tax Court: “JUST DO IT.” If predictive coding is a good fit for the case because it will save time and money, don’t file a motion seeking preemptive permission from the court, JUST DO IT. Try to seek agreement from the other side on a defensible predictive coding protocol, but if they refuse to act reasonably, JUST DO IT. Or if you know your obstreperous opponent will not agree to a protocol, but will instead turn your predictive coding proposal into a discovery weapon, don’t even suggest an agreement, JUST DO IT.
Of course, to employ the “JUST DO IT” strategy, the litigant must accept the risk that it will have to defend its decision in court, and that it may be forced to re-produce documents using another method. It must ensure that its predictive coding methodology is defensible – reasonably designed, well documented, statistically tested and validated, and consistent with best practices for the software being used. And the litigant should be mindful of any representations already made about how discovery will be accomplished – as we saw in the Progressive Casualty Insurance case earlier this year, a party that commits to producing documents using a certain protocol cannot later change the protocol unilaterally.
Our goal is to provide DiscoverReady’s clients with tailored discovery strategies for their uniquely individual matters. For some clients, JUST DO IT will be the right approach. For others, probably not – although we might work with those clients to find other, more conservative ways of leveraging predictive coding, combined with traditional methodologies, to reduce the cost of discovery. (For example, predictive coding could be used to prioritize review and more efficiently deploy resources; or it could be used at the tail end of a prioritized, manual review to support a request to stop further review and production.) In the end, all clients share the same objective, recognized by Judge Buch in his Dynamo order: “secure the just, speedy, and inexpensive determination of every case.” As more and more courts endorse predictive coding, that objective should be easier to achieve.
A recognized thought leader in e-discovery, Maureen collaborates with the company’s clients and operations teams to develop innovative information strategies for legal discovery, compliance, and sensitive data protection. She speaks and writes frequently on significant issues in e-discovery and information governance, and participates actively in the Sedona Conference Working Groups on Electronic Document Retention and Production and Data Privacy and Security. Prior to DiscoverReady, Maureen was a partner at Paul Hastings LLP, where she represented Fortune 100 companies in complex employment litigation matters.