My point in this column is that metrics predominate in e-discovery generally and especially for one particular category of software. Analytical discovery software determines the relevance of documents in a collection based on the contents of the documents rather than on the presence in them of key words or phrases. Sometimes referred to as machine-aided review, the software gathers similar documents into clusters, makes binary relevance determinations or ranks the documents they find according to an algorithm for relevance.
Metrics rule in e-discovery. Generally speaking, those who manage e-discovery scope their tasks with metrics. So many terabytes of data collected from so many custodians, such and such a percentage of relevant documents compared to a lower percentage of privileged documents, so many documents reviewed per hour per person. Keyword searches report how many documents had hits. At every step, therefore, averages and medians and percentages are the vocabulary for describing how much time the process takes, what's involved and what the likely expense will be.