In this episode of The Digital Detectives, Sharon Nelson and John Simek discuss the evolution of e-discovery, how predictive coding works and is priced, and examples of cases that have involved predictive coding.
Simek first explains the importance of culling, or filtering out unimportant data sets through DeNISTing, deduping, or filtering by dates. He then explains predictive coding in its simplicity: to feed a computer program information based on discovery attorneys have already done until the computer can accurately predict which information is important.
Simek and Nelson then go on to examine the prices vendors charge for the predictive coding process and in which cases it might be profitable for the law firm or client. There is a steep, expensive learning curve involved; many mid-sized law firms probably will not profit and even very large cases only save an average of 15% using predictive coding. However, Nelson explains, predictive coding is the future of discovery, so it is important for lawyers to pay attention to when the benefits outweigh the costs.