Using Technology to Reduce the Cost of Disclosure

Electronic-disclosure (also known as e-disclosure) has been widely used in English litigation for many years, but the process may be streamlined following a recent decision of the English High Court. We look at a recent case where the High Court has enthusiastically supported the use of ‘predictive coding’, which makes greater use of technology in the review process.

29 February 2016

One of the major costs of any litigation case has always been the review and disclosure of documents in each party’s possession. With new methods of electronic communication and digital data storage becoming much more widely used, the volume of documents being detected which might be of relevance in litigation is increasing. As a result, the cost of reviewing these materials to ascertain whether they should be disclosed is also rising.

However, a recent decision of the High Court (Pyrrho Investments Limited & Anr v MWB Property Limited and Others [2016] EWHC 256 (Ch)) could represent a new dawn for litigators as the case clearly demonstrated that ‘predictive coding’ technology could reduce the current labour intensive and expensive review process.

Typically, documents identified as potentially relevant to the litigation are reviewed manually, to ensure that only those that are relevant to the issues at hand are disclosed to other parties. The traditional methodology has been to use large numbers of junior legal staff to look at each document and indicate whether it falls within the parameters of disclosure in the case at hand. However, even with the rise of off-shoring or ‘Northshoring’ (the outsourcing of work to lower cost areas such as the north of England), this process remains cumbersome and expensive.

Predictive coding offers an alternative. It seeks to simplify, accelerate and reduce the cost of disclosure exercises while maintaining, or indeed improving, accuracy.

What is predictive coding?
Predictive coding, also known as technology assisted review (TAR) involves a three-step process.

  1. Firstly all the documents that may be relevant to a case are collected into one dataroom. A senior lawyer (or small group of lawyers) (the expert reviewer(s)) with knowledge of the case will then manually review a small sample of these documents and categorise them based on relevance.  There may be several rounds of quality control within this process. This process is used to ‘train’ the software.
  2. Next, the predictive coding software identifies those aspects of the sample documents that the expert reviewer has highlighted as important and finds similar patterns in the remaining documents contained in the dataroom.
  3. Finally, the program will score the documents based on the factors identified in the expert review to reflect relevance. Documents that achieve a certain score can then either be immediately disclosed or put through a traditional manual review. For further details of the process, please refer to our previous article: Focus on Predictive Coding

Use of predictive coding in English litigation
Despite the availability of such technology, litigation parties have appeared to be hesitant to use it. This hesitation should now have been largely dispelled due to a strong endorsement of predictive coding in the recent case of Pyrrho Investments Limited & Anr v MWB Property Limited and Others [2016] EWHC 256 (Ch).  In this case, the court acknowledged the “novelty” of the use of this kind of software in English proceedings and the lack of existing case law in this area.

The case concerned a joint application by the parties for approval to use predictive coding. The court granted the application and gave strong approval for the use of the software in its judgment. In particular it highlighted that:

  • Predictive coding has been accepted in other jurisdictions such as Ireland and the USA.
  • Nothing in the Civil Procedure Rules (CPR) prohibits its use.
  • There was no evidence of predictive coding being less accurate that manual review.
  • Greater consistency would be gained as predictive coding replicated the analysis of a single senior lawyer (or select group) as opposed to a wide range of junior lawyers.
  • While predictive coding technology could be expensive, it may well be far less expensive than full manual review.

Following this case, it is likely that the use of predictive coding will become more widespread in English litigation. As in many cases the use of this technology will reduce the costs for litigants, this move must be welcomed. 

Shepherd and Wedderburn have a depth of experience in using technology assisted review, including predictive coding, and have a dedicated e-discovery team who can help you with any queries you may have about predictive coding and its potential uses. Please do not hesitate to contact John Mackenzie, Claire Stockford or Nicola Perry if we can assist you any further on these matters.