Following the first post in this series, I got some excellent feedback the six readers (not including my mother) that read it. So, buoyed by the encouragement of my global readership, let me move right on to the next pain point shared by eDiscovery managers: the challenge of risk and risk assessment.
There are two key risks in eDiscovery – defensibility and winnability. Let’s take defensibility first -- a key risk in eDiscovery centers around the retrieval and production of responsive documents. This, after all, is the key objective of the eDiscovery process, and the failure to properly discharge this central function is often the fundamental axis around which assessments of risk revolve. But here is the paradox. The most widely used practice in eDiscovery, the practice on which there is universal consensus, and whose defensibility is unquestioned, is keyword search. Keyword matches are the main vehicle for culling the raw collection ingest, and in preparing the review set. Yet keyword search is broken. As demonstrated in a growing series of studies, most recently in the TREC 2008 experiment, keyword matches retrieve in the region of 20 to 30% of the relevant documents. And, this is the gold standard of defensibility in eDiscovery. There are, however, growing signs that this paradox, which has been lying hidden at the heart of the eDiscovery process, is progressively being exposed.
Winnability (or the lack of it) also represents a huge risk, but one that strangely receives much less attention than its defensibility counterpart (in fact, winnability has attracted so little attention that, as far as I aware, it has not, until this post, merited its own terminology). I would suggest that the reason for this is that until very recently, winnability has been a problem without a solution. As technology vendors know, there’s no point analyzing the problem if you don’t the technology that can solve it. Mitigation of the risk of winnability would be afforded by a technology that enabled fair assessment of your chances of winning a case. It’s only recently, with the emergence of early case assessment as a recognizable practice in the litigation cycle, that practitioners have had at their disposal a body of methodology and technology that address the problem. | 
