Chancery Division of the High Court approves the use of "predictive coding" in an e-disclosure exercise for the first time
This is a landmark decision of the High Court, approving the use of predictive coding for the first time. "Predictive coding" (also known as technology, or computer, assisted review), is a method whereby software analyses documents and "scores" them for relevance, and thereby reduces both the time and costs needed to complete an electronic disclosure exercise. Typically, the parties agree a protocol and a representative sample of potentially relevant documents is then obtained.
In this case, the parties had agreed the use of predictive coding and sought approval for this from the court. Other than a fleeting reference to predictive coding in one earlier case, there has been no other consideration of this issue by the English courts. However, other jurisdictions (such as the US and Ireland) have endorsed the use of this software.
Master Matthews gave his approval to the parties to use predictive coding. He noted that nothing in the rules/practice directions prohibits its use and there is no evidence that it leads to less accurate disclosure and there is some evidence to the contrary.
In this particular case, a full manual review of over 3 million documents would be unreasonable. Predictive coding would also cost far less than the full manual alternative. Since the "value" of the claims made in this case runs to tens of millions of pounds, the estimated costs of using the software were proportionate to the value of the claim. It was also of relevance that the trial would not take place for over a year, and so there would be plenty of time to consider other disclosure methods, if that became necessary. The Master further noted that the parties had agreed the use of the software and that such use would be appropriate here and would promote the overriding objective.
There would be greater consistency in using the computer to apply the approach of a senior lawyer towards the initial sample to the whole document set than in using numerous lower-grade fee-earners, each seeking independently to apply the criteria. The Master advised that "best practice" would be for a single, senior lawyer, who has mastered the issues in the case, to consider the initial representative sample (marking it as relevant or not), in order to "train" the software to review the whole document set. Further statistical sampling by humans must then be conducted to ensure the quality of the exercise. Once an acceptable level of accuracy is reached, the software then categorises all the documents.
The Master did caution, though, that: "Whether it would be right for approval to be given in other cases, will, of course, depend upon the particular circumstances obtaining in them."