ECA is a type of triage process for accelerated e-discovery at the beginning of a case. It focuses on the discovery/location, analysis, and review of electronic documents (ESI) that are highly relevant. The documents so located are typically referred to as the “low hanging fruit” and this is not intended to be a complete search and review. The targeted hot documents found in the ECA process allow trial attorneys and parties to begin to immediately evaluate the merits of the case and serve as initial guides to subsequent case handling, including witness interviews and other investigations. The simply relevant and highly relevant documents located in an expedited ECA allow e-discovery specialists to:

1) guide ongoing ESI identification, preservation and collection efforts;

2) refine initial search strategies;

3) begin to develop final review and production strategies; and significantly;

4) to eliminate, or cull out, large segments of ESI collected that is unlikely to have relevant information and is thus removed from further review;

5) refine initial projections and estimates of e-discovery costs, especially review costs, and formulate realistic budgets, and proportional scope strategy (the last item I like to call Bottom Line Driven search and review under 26(b)(2)(C)).

The highly relevant documents are located through various expedited search and review procedures, including:

1) manual review with judgmental sampling and random sampling using native software, i.e. – review of a custodian’s email through their PST file looking at their email collection the same way they see it;

2) evaluating the mental set of custodians by the native sampling;

3) testing and refinement of keywords;

4) utilization of special software designed to facilitate and enhance the above processes and include new search techniques available in the particular software, generally known as concept search methods. The concept search functions in most software includes:

a) visualization of witness and ESI relationships;

b) search of concepts related to key words and ideas;

c) location of documents with a percentage of similarity to documents identified by you as relevant, usually on a sliding scale by percentage; and,

d) various clusters of documents that the software recognizes and then presents for determination in bulk as to relevance (for example, you identify one representative type of document as irrelevant, and the software automatically marks all similar document as irrelevant, such as sports communications).

The ECA allows for the intelligent and legally defensible culling of most of the ESI in a collection, typically 90%, including processing culling such as deduplication and deNISTing. (My personal goal in all cases is to always cull out 98% in a legally defensible manner, not just the industry standard of 90%, and I believe this is attainable in many cases.) The ESI that is not culled out by ECA, or processing, is then passed through for actual full review, not just partial judgmental and random sampling review. Documents specifically identified as highly relevant or relevant in ECA will retain that tagging, but will be subject to further review for possible privilege, confidentiality, or sometimes limited issue tagging.

via What is ECA? Here is my definition. « e-Discovery Team.