The District of Nevada has held that a party improperly used predictive coding, and should produce all documents containing keyword hits, in Progressive Cas. Ins. Co. v. Delaney, No. 2:11-cv-00678-LRH-PAL (July 18, 2014) (amending Order of May 19, 2014). Although the court rejected use of predictive coding in this case, the lesson for future litigants is simply this: Don’t agree to a discovery protocol if you have not made efforts to understand the potential burdens associated with it. Here, the plaintiff agreed to use keywords, and was held to that burden even when it turned out to be very expensive. In the court’s words, “Had the parties worked with their e-discovery consultants and agreed at the onset of this case to a predictive coding-based ESI protocol, the court would not hesitate to approve a transparent, mutually agreed upon ESI protocol.”
Defendant FDIC as receiver requested documents from the plaintiff pursuant to a joint ESI protocol stipulated by the parties. The protocol provided that documents would be identified according to a keyword search, and could be reviewed or simply produced without review. A protective order permitted clawback of any privileged documents.
After agreeing to this protocol, the plaintiff realized the burdensomeness of complying with it. Rather than review the 565,000 documents with keywords, Plaintiff instead determined it preferred to use predictive coding to review the documents with keywords. Plaintiff did not include the documents without keyword hits in its predictive coding review. The FDIC challenged this approach, arguing that it violated the stipulated ESI protocol. The court agreed.
While the FDIC’s argued that Plaintiff’s predictive coding methodology created invalid results, the court’s holding was based on Plaintiff’s lack of cooperation and transparency. Plaintiff had agreed to an ESI Protocol based on keywords, and it was not entitled to deviate from that without seeking agreement from the FDIC. The court ordered Plaintiff to produce all documents without further review, except documents containing keywords reflecting that they might be privileged.
Parties often feel trapped by discovery burdens that are of their own making. Once a party agrees to a protocol, it has committed to the burdens associated with that protocol. Therefore, a party should attempt to determine the potential costs a search method before agreeing to it. Guessing at keywords (sometimes called the “go fish” method) without testing them often leads to a high volume of false hits, as well as many missed documents.
For party who finds it needs an ordered ESI protocols amended, the court is more likely to be sympathetic if the party has first attempted to cooperate with the other side. It should should be prepared to prove that the burden is not proportional to the case, and attempt to reach agreement with the requesting party on a more appropriate protocol. Still, it is far better to go through this process while setting the ESI protocol the first time, rather than try to amend it. Plaintiff appears to have understood this risk, as it refused to share information with FDIC about its new proposed protocol because it believed that “short of throwing open its doors to the FDIC, both physical and electronic, and allowing the FDIC and its counsel to rummage through its files at will, there is nothing Progressive can do with respect to discovery that will not lead to still more complaints from the FDIC, still more demands for additional documents and information, and still more motion practice.”
 FDIC argued that it “would not have narrowed its search terms so significantly had Progressive revealed it intended to overlay predictive coding onto the narrowed set.”