The Neutral Corner: Recapturing the Benefits of Predictive Coding by Using an eDiscovery Special Master

Topics: ediscovery, Efficiency, Law Firm Profitability, Law Firms, Legal Innovation, Midsize Law Firms Blog Posts

predictive coding

Following a New York federal court’s approval of predictive coding in the Rio Tinto case, attorneys are increasingly using predictive coding to expedite discovery when faced with large quantities of documents.

While predictive coding can save significant time and money in discovery, parties often disagree on the methods and protocols for implementing it, potentially negating the time and cost savings. Courts or the involved parties may want to push for the appointment of an eDiscovery Special Master as early in a case’s lifecycle as possible to mediate protocol disputes and provide guidance on technical issues, hopefully recapturing the benefits of predictive coding and saving the parties time and money.

What is Predictive Coding?

Predictive coding is, essentially, the use of keyword search, filtering and sampling to “teach” a computer to identify responsive or privileged documents. While there are multiple variants, the basic process is as follows. First, a set of documents, referred to as the “Seed Set” is pulled from the entire corpus of documents for review. A subject matter expert, intimately familiar with the case, then codes the Seed Set for responsiveness and privilege. A computer analyzes the expert’s coding, attempts to learn what makes documents responsive or privileged, and codes a new set of documents. The expert then corrects the computer’s mistakes, and the computer is tasked with coding another set of documents, which are then reviewed by the subject matter expert. This process is done iteratively until the computer can hit certain statistical benchmarks. Once the computer is properly trained, it codes the entire document corpus.

Predictive Coding Processes & Protocols Can Cause Substantial Disputes

Unfortunately, predictive coding processes and protocols are breeding grounds for disputes. One of the challenges facing parties is related to implementing and monitoring compliance with stipulated predictive coding processes and protocols.

In Rio Tinto,for example,the court approved the parties’ stipulated “Predictive Coding Protocol” that was supposed to make the use of predictive coding self-executing. In this case, execution of the protocol soon proved to be a major point of contention, and the parties sought judicial relief on several issues. Specifically, the court was asked to resolve a number of technical issues related to interpreting the Predictive Coding Protocol, including what search terms were to be used and challenges between the parties on issues of training, disclosure and the sufficiency of the predictive coding process itself. While the court resolved the issues, the motion practice and hearings were both time consuming and expensive and involved extensive argument between the parties’ experts.

How Would a Special Master Have Helped

Appointing E-Discovery Special Masters early in the case lifecycle can help avoid and mitigate predictive coding disputes such as those in Rio Tinto. (Note: The court eventually did appoint a Special Master after months of trudging through issues with the protocol, but by that time, the litigation was already significantly disrupted.)

Special Masters can use their technical knowledge to create unambiguous provisions that will reduce disputes downstream. As noted above, there is not one correct way to do predictive coding, which means that a Special Master can provide insights into methods for selecting search terms, seed sets, precision and recall standards that are mutually agreeable to the parties and fit their technical capabilities.

Additionally, Special Masters with the requisite technical expertise can monitor compliance with predictive coding protocols and make judgements on issues that may be beyond the court’s or parties’ understanding. If done correctly, the Special Master eliminates the need for the parties to have their technical consultants argue their positions before the court, which saves time and money. Finally, the Special Master’s presence as a neutral technical expert can assist the parties in finding common ground on technical issues that otherwise would have become contentious due to the parties’ inexperience with the technology.

Predictive coding is useful technology and has potential to save time and money. While Special Masters themselves cost money, often the benefits they provide allow for a net gain in using predictive coding that may not be realized if the court and parties are left to their own machinations.


Michael Mann, a Senior Electronic Discovery & Forensic Consultant at Law & Forensics, contributed to this blog post.