The E-Discovery Field Guide

Court Blesses Use of TAR without Prior Court Approval

In this month’s edition of the E-discovery newsletter we break from our norm of providing a few shorter topics to address one case at length. This recent decision involves use of technology-assisted review (TAR). TAR is a process of having computers classify documents based on input from document reviewers. For example, responsive, privileged, or other categories of documents can be identified from a massive set of data via TAR based on a smaller “seed” set of representative documents. TAR works by searching and finding documents that are similar to this initial seed set. Advances in technology and computing power have made TAR a viable review tool for all but the smallest ESI datasets.

Recently, the use of TAR as a tool to review electronic data has increased in both use and acceptance. Unsurprisingly, as the use of TAR has increased, the number of court decisions addressing TAR has correspondingly increased. This process has created a positive-feedback loop: as courts become increasingly comfortable and accepting of TAR, parties have more confidence in using TAR as a tool.

Clear court acceptance of use of TAR: Entrata, Inc. v. Yardi Sys., Inc.

Entrata, Inc. v. Yardi Sys., Inc., No. 2:15-cv-00102, 2018 WL 5470454, (D. Utah Oct. 29, 2018), stands out as one example of a court blessing a party’s decision to use TAR.  The decision provides strong support for parties to use TAR in the future. And although it addressed use of TAR specifically, Entrata contains many lessons for ESI discovery in general.

Entrada begins, as many cases do, with the parties meeting to discuss a discovery plan and submitting that plan to the court. See Fed. R. Civ. P. 26(f). Following that submission, the parties discussed Entrata’s proposed use of TAR. Yardi’s counsel sent Entrata a list of questions about the proposed TAR process and Entrata’s counsel answered these, including: how Entrata would identify the “seed” set of documents; whether Entrata would filter documents before use of TAR; and how it would handle documents that could not be categorized by TAR. Entrata’s counsel said that it would use TAR as a “culling” tool and would then conduct a “linear” review of the documents that TAR identified as responsive.

During this conferral process, the parties submitted, and the court later ordered, an order governing the ESI discovery. The order provided that the parties should work together to “identify and negotiate a reasonable set of search terms or other search methodology to be used in searches of ESI.” If the parties could not reach agreement within thirty days of the order, the parties would submit competing proposals.

Neither party submitted a proposal under this order.

Yardi, however, later raised issues with Entrata’s use of TAR and requested specific information relating to its TAR process. On the last day of fact discovery, Yardi filed a motion to compel seeking production of TAR information. Entrata opposed, arguing that Yardi had waited too long to file the motion and failed to offer any reason why the TAR process was deficient. The magistrate judge found that there were no specific reasons given why the process was deficient and that, if Yardi had any such concerns, it should have addressed those with the court long ago.

Yardi objected to the magistrate’s order and a flurry of briefing before the district judge commenced. Yardi argued that the magistrate’s ruling was clearly erroneous because the Federal Rules required Entrata “in the first instance to provide transparent disclosures as a requirement attendant to its use of TAR in its document review.” The district court rejected this argument.

The court first addressed Rule 26(g), noting that there is nothing in that rule that “obligates counsel to disclose the manner in which documents are collected, reviewed, and produced in response to a discovery request.” Yet, the court said, Rule 26(f) does require the parties to confer about the ESI discovery process. The court noted that 26(f) was not merely a “perfunctory exercise” but a meaningful opportunity for the parties to address, upfront, ESI issues before they escalate. These issues, according to the court, must be revisited as the case progresses. The district court summed up the parties’ ESI obligations noting that the “Federal Rules of Civil Procedure assume cooperation in discovery.” The parties, however, reached no such agreement. Yardi knew, nonetheless, of Entrata’s use of TAR and failed to raise the issue with the court until the very last minute. In sum, too little, too late.

The court also rejected Yardi’s argument that Entrata had to provide TAR disclosures as a matter of course. The court distinguished the cases Yardi supported this argument with noting that they all involved TAR processes which the parties had agreed upon. The court said that it was “black letter law” that courts will permit a producing party to use TAR and Entrata did not need to seek approval from the court for use of TAR. If Yardi had concerns about the use of TAR, it should have raised them well before the final day of discovery.

Finally, the court rejected Entrata’s request for attorney’s fees, finding that Yardi’s position was not wholly meritless.

Implications 

Entrata is notable for its endorsement of TAR as an ESI review method. As the court said, it is now “black letter” that TAR is an appropriate method to identify responsive documents. Courts have gained such familiarity and comfort with the TAR process that it is no longer viewed as a “new” technology requiring court approval and supervision. TAR can now be viewed as a viable ESI tool, like search terms, that a party can employ to identify and produce responsive ESI with confidence.

There is, however, a larger lesson in Entrata, one applicable to all types of ESI discovery: courts expect cooperation in discovery and, if agreement cannot be reached, the ESI dispute should be brought to the court’s attention as early in the discovery process as possible.

Entrata was perhaps loath to disclose the nuts and bolts of its TAR process to the other side for fear that it would lead to the other side’s objecting to its selected process. Entrata’s failure, however, to get the other side’s approval or the court’s endorsement led to a long and expensive discovery dispute (one brought at the end of discovery and for which victorious Entrata did not recoup its costs). If Entrata had disclosed its TAR process to the other side or sought court approval of this process, it may have, ultimately, foreclosed the larger fight it had at the end of fact discovery. Perhaps the dispute could have been avoided entirely. Or maybe the dispute would have been a skirmish, not the full battle that ultimately resulted. But Entrata never found out how much of the TAR process, if anything, would have been disputed.

Whatever the method of review for ESI, parties are often hesitant to disclose to the other side the exact review process (i.e., search terms, custodians, TAR processes) for fear that that the opposing party will simply pick it apart and a needless dispute will arise. But, as shown in Entrata, there are real risks to keeping an ESI process in the dark and having it challenged late in the game. Courts now expect cooperation in ESI discovery and failure to engage in this process has consequences to a party, even if there are no substantive problems with a party’s ESI production. The better approach is to seek agreement first and, failing such, to stake out a defensible ESI position with the court as early as possible in the discovery process. At least this much is clear: unlike Yardi, don’t wait until the very last day of discovery to bring an ESI dispute to the court for the first time.