Developing Effective Key Words is an Iterative and Thought Intensive Process
June 27, 2018
In my December 2016 blog post, I wrote about how developing effective key words is very much an iterative and thought intensive process. This message was recently reaffirmed by a decision out of the Southern District of Ohio, wherein the Judge reminded us that the process of identifying search terms it not merely one of guesswork. Rather, it requires deliberative thought, testing and refining such that the searches yield target-rich and proportionate results.
In the complex commercial matter, AMERICAN MUNICIPAL POWER, INC., Plaintiff, vs. VOITH HYDRO, INC., Defendant, the parties were in litigation over the construction of four hydroelectric power plants (“Projects”) that spanned a decade.
When it came to discovery, the parties could not have been more oppositional in their respective stances. Voith Hydro, Inc. (“Voith”) requested plaintiff conduct a search for ESI using only the four Project names (Cannelton, Smithland, Willow and Meldahl) as standalone search terms without qualifiers.
American Municipal Power, Inc., (“AMP”) on the other hand, requested that Voith search its ESI collection without reference to any of the four project names; searching instead for various employee and contractor names connected in a boolean way with a list of common construction terms and the names of hydroelectric parts restricted only by date range.
However, both proposals were too expansive in their scope and, as a result, too burdensome. For example, when proposing its search terms, defense counsel did not appreciate that not every shred of ESI that hit upon one of the four fairly common project names (i.e., Willow) would be relevant or, at the very least, worth examining for relevance. As the Court noted, “As owner, AMP had retained millions of documents for more than a decade that contain the name of the four projects which refer to all kinds of matters unrelated to the case.” The Project names –detached from any other qualifier or restriction – returned documents related to real property acquisitions, licensing, employee benefits, facility tours, parking lot signage, etc. All content that “hit” on a search term, but none of which was relevant to the construction of the four hydroelectric power plants. Thus, accepting Voith’s request would result in overly broad results not sufficiently tailored to the needs of the case. And so, Judge Deavers ruled against the defense on the four project names keywords request, and granted a protective order for the plaintiff because, in the Judge’s words:
“The burden and expense of applying the search terms of each Project’s name without additional qualifiers outweighs the benefits of this discovery for Voith and is disproportionate to the needs of even this extremely complicated case.”
AMP’s more “expansive” search terms, however, were equally problematic. Indeed, if the searches combined employee and contractor names together with a list of common construction terms and the names of hydroelectric parts, the results included innumerable confidential communications about other projects Voith performed for other customers. Therefore, like Voith’s request, AMP’s search request similarly returned a large corpus of documents not relevant to the litigation. AMP’s proposed searches also raised confidentiality concerns (i.e., commercial information about other customers).*
Keyword searching should be an iterative process that is tested and refined, and tested and refined again, before any commitments to search are made.
* AMP proposed to exclude the names of other customers’ project names with “AND NOT” exclusionary phrases. This was deemed by the Court unworkable because Voith could not possibly reasonably identify all the projects from around the world with which its employees were involved during the decade they were engaged in work for AMP on the Projects.
Have questions on keyword searching? Please contact me at email@example.com.