12 March 2018Print This Post

City firm foresees big future for predictive coding after client wins case

Glynn-Jones: Predictive coding effectively deployed

The City law firm that won the first contested application to use predictive coding as part of a substantial document review exercise has proclaimed the exercise a success and a precursor to its wider use after its client won.

Berwin Leighton Paisner (BLP) said it proved particularly effective in a case like this, where the nature of the dispute “did not allow the formulation of a sensible set of key words for a traditional disclosure review”.

Brown v BCA Trading saw the petitioner David Brown claim over £20m from the respondents through an unfair prejudice petition brought under Section 994 of the Companies Act 2006.

Mr Registrar Jones made the order for electronic disclosure using predictive coding, rather than the “more traditional” keyword approach, in 2016.

He said the cost of predictive coding was in the region of £132,000, compared with at least £250,000 for a keyword search or £338,000 in a worst case scenario.

BLP reported today that the documents uncovered by the search directly disproved a number of the petitioner’s allegations, and cast doubt on other aspects of his evidence. The petition was dismissed last week after a 12-day trial, with BCA awarded its costs.

A briefing from BLP said: “The predictive coding disclosure exercise undertaken for BCA proved to be a success. The petitioner, having resisted it throughout, could find only minor ways to challenge its output, resulting in discrete further searches being performed by BLP.

“Predictive coding was particularly effective in a case such as Brown v BCA, which contained broad allegations of unfair prejudice and bad faith. These issues did not allow the formulation of a sensible set of key words for a traditional disclosure review.

“Instead, the computer was able to ‘learn’ from the sample reviews performed by senior BLP lawyers in order to ‘educate’ it.

“The predictive coding algorithm that was created as a result of the human review and refinement process was able to produce an impressive number of documents that were highly relevant to the issues, many of which would be unlikely to have been retrieved using a proportionate set of key words in a traditional process.

“While the converse is also true, i.e. that the predictive coding algorithm did not produce all of the same documents that would have been retrieved using key words, and predictive coding does not aspire to perfection, it is believed to achieve results that are superior overall to a traditional review, at a much reduced cost to the client.”

Oliver Glynn-Jones, BLP’s head of commercial dispute resolution, said: “We are delighted with the successful outcome of this case, both as a positive outcome for a valued client and as evidence of how predictive coding can be deployed effectively during litigation…

“Now that the technology has been tested and proven at full trial, and demonstrated benefits in terms of cost and accuracy, we predict that it’s likely to become much more prevalent in commercial litigation.”

By Neil Rose

Leave a comment

We encourage you to be part of the Litigation Futures community but please note that all comments will be moderated before posting. We draw your attention to clause 5 of the Terms and Conditions of the site, which deals with user-generated content.