Making Sense of ‘Kleen Products’: Is it Really about Predictive Coding?

In the wake of Magistrate Judge Andrew J. Peck’s ruling in Da Silva Moore v. Publicis Groupe affirming the use of predictive coding, many in the e-discovery field turned their attention to Kleen Products LLC v. Packaging Corporation of America, believing that it might be the Next Big Case on predictive coding. The plaintiffs in Kleen Products have asked U.S. Magistrate Judge Nan Nolan to require the defendants to use predictive coding, even though defendants have already produced more than 3 million pages of discovery using more traditional methods of Boolean search. Judge Nolan conducted two days of evidentiary hearings on the request in February and March and has subsequently held several status conferences with the parties.

To date, Judge Nolan has issued no formal ruling on the issue. Instead, she has encouraged the parties to attempt to come to terms on their outstanding discovery issues. There is a soft deadline looming in the background over the dispute, in that Judge Nolan is slated to retire on Oct. 1, 2012.

Last month, Matthew Nelson posted insightful observations about the case at the Clearwell blog e-discovery 2.0,  together with transcripts of the March 28 hearing and subsequent status conferences. In what he describes as “big news for litigators,” he notes that Judge Nolan has urged the parties to focus on developing a mutually agreeable keyword search strategy. In addition to the transcripts he posted, I also found the transcript of the Feb. 21 hearing at the website for a University of Maryland e-discovery seminar taught by Doug Oard and Jason Baron.

The Predictive Coding ‘Detour’

Having heard so much about this case in the context of predictive coding, I started to make my way through these transcripts for clues as to how it might turn out. The more I read, the less this seemed like a predictive coding case at all.

Granted, much of the testimony and colloquy centered on predictive coding and the adequacy (or inadequacy) of keyword search. But if you focus on Judge Nolan’s remarks, she seems to be doing everything in her power (short of issuing an order) to divert the discussion away from predictive coding. As a matter of fact, at one point in the March 28 hearing, she wonders how the case got on the “detour we’re on with predictive coding,” and refers to the fact that “all the blogs are talking about us.”

Her comments reveal that her overarching concern is not technique but results. She makes this clear immediately after her detour remark, when she says, “I assume … what you really are interested in is a search, regardless if it’s Boolean or computer-assisted, that is fair … and that can be validated statistically.” And the best way to achieve results that everyone will consider fair, she seems to suggest, is through cooperation.

Near the end of the March 28 hearing, she issues the parties a “homework assignment” to talk to each other and see if they can come up with “tweaking” to the searches that had already been done that would provide the kind of validation the plaintiffs are seeking. She gives no indication that she is inclined to order the defendants to start from square one with predictive coding.

When the parties come back before Judge Nolan on April 19 for a status conference, they report that a series of meet-and-confer sessions had failed to resolve their issues. Once again, she tries to turn the focus away from technique and toward results:

[W]hat I learned from the hearing, what seemed to be the most important is that whatever your method would be, whether it would be boolean or predictive coding or whatever they are going to call it next year, is that the parties and to a certain extent the Court, but I think it’s more the parties, want to feel and be able to say that this is a verifiable, somewhat verifiable and accurate method that they have chosen.

Sensing that some progress had been achieved through the meet-and-confer sessions, the judge urges the parties to address open issues in another round and then come back to her. That return session was scheduled to be held on May 22. I cannot find any reports indicating whether it went forward that day or, if so, what happened.

A Case About Cooperation

Throughout the transcripts, Judge Nolan makes one point repeatedly. During the few months remaining before her retirement, the parties can work with her to resolve their discovery disputes or they can “be coming in on motions to compel on this for the next two years to whoever the new magistrate judge is.”

On paper, Kleen Products may look like a case about predictive coding. But as you read through these transcripts, it becomes clear that it is really a case about cooperation. When parties are able to cooperate and agree in e-discovery, Judge Nolan seems to believe, then they are more likely to be satisfied with the outcome — regardless of what they call the process they use to get there.

4 thoughts on “Making Sense of ‘Kleen Products’: Is it Really about Predictive Coding?

  1. Pingback: Catalyst E-Discovery Search Blog: Making Sense of ‘Kleen Products’: Is it Really about Predictive Coding? - ADR Toolbox

  2. Pingback: The Next Big Predictive Coding Case that Wasn’t

  3. Pingback: Technology-Assisted Review: From Expert Mentions to Mainstream Coverage | @ComplexD

  4. Pingback: Technology-Assisted Review: From Expert Mentions to Mainstream Coverage | @ComplexD

Leave a Reply

Your email address will not be published. Required fields are marked *