All around the world, good lawyers seek increasingly efficient eDiscovery strategies. It is thus no surprise that the UK, father of our own common law dating back to the Magna Carta, has joined the US and Ireland in accepting predictive coding (see sidebar). In Pyrrho Investments Ltd. v MWB Properly Ltd., U.K. Master Mathews ruled that predictive coding was permissible for ten reasons which, paraphrased, are:

  1. Other jurisdictions, such as the US and Ireland, have ruled that predictive coding can be useful in certain cases;
  2. Predictive coding leads to accuracy that is equal to or better than manual review;
  3. Predictive coding yields greater consistency than large teams of junior lawyers, each applying their own interpretation of the coding protocol;
  4. There is nothing in English law prohibiting predictive coding;
  5. In the instant case, there is a huge (over 3 million) collection of electronic documents;
  6. and 7. The cost of manually reviewing the full document collection would be “unreasonably” high given the lower cost estimates of using predictive coding;
  7. The case is valued in the tens of millions of pounds, so the cost of using predictive coding software is proportionate;
  8. There is sufficient time before trial to consider other methods of review if the predictive coding turns out to be unsatisfactory;
  9. The parties agreed to the use of predictive coding.

He concluded, “Whether it would be right for approval to be given in other cases will, of course, depend upon the particular circumstances obtaining in them.”

The court’s decision referenced U.S. Judge Peck’s seminal Da Silva Moore v Publicis Groupe & MSL Group decision and the Irish High Court endorsement of predictive coding in Irish Bank Resolution Corporation Ltd. v Quinn as persuasive, albeit non-binding, precedent.

Interestingly, although the case comes just less than a year after U.S. Judge Peck’s latest opinion on predictive coding in Rio Tinto Plc v. Vale S.A., it provides no reference to that ruling. It is, however, consistent with Judge Peck’s continuing theme that where both parties wish to use predictive coding, there is no reason for courts to interfere or dissuade them. Modern litigation technology can be trusted, and is indeed a beautiful thing.

We at Elevate believe it is just a matter of time before Australian courts follow suit.

About the Author

Marilyn O. Primiano serves as Director of Litigation for Elevate Services, where she focuses on ensuring quality and efficiency for litigation and compliance matters. Her significant experience with law firms and in legal project management inform her consultative legal project guidance to companies and law firms worldwide. Marilyn has lived and worked in India, Europe, and the US.

Predictive coding

is a tool within the technology assisted review (“TAR”) spectrum, although the two terms are often incorrectly used interchangeably. TAR includes everything from common keyword searching to email threading. Predictive coding is a technology which auto-codes documents as likely responsive or non-responsive based upon machine learning, typically acquired though senior attorney coded “seed sets” or “alpha sets.”


“Modern litigation technology can be trusted, and is indeed a beautiful thing.”

Share This