I recently wrote that choosing onshore v. offshore review is an empirical question. Here’s some good reading to help think about answering this. 

In Onshore v Offshore Pivot Point – Part II I said in passing that software might be an even better choice for document review than lawyers, onshore or off. “Of Litigators and Butterflies: The Quest for a Quantum Leap in Large-Scale Document Review” in Pike & Fisher’s Digital Discovery and e-Evidence (July 2006) and available at H5 Technologies’ web site as a PDF is a good analysis of the inherent error in human document review.

Author Nicolas Economou, H5’s CEO, sets forth a mathematical equation for assessing document review error. Though the equation is conceptual for now, he also presents provocative empirical data. A study found that the H5 automated approach correctly identified relevant documents 98% of the time.

Of course, many lawyers will immediately react that this is simply unacceptable – how can we live with a 2% error rate. Those lawyers, however, most likely lack empirical data about the accuracy of their army of lawyers.

The article also reports that “for each relevant document missed by the H5 system, the control review process missed 32 documents. That is, the risk of failing to flag relevant documents for litigator review was 32 times greater under the traditional review process.”

If this is true and can be further substantiated, at what point will lawyers commit malpractice by continuing to rely primarily on human review?