Published on Sep 29, 2016 
Rarely does anything come easily in e-discovery and negotiating a protocol regarding the use of technology-assisted review (TAR) with a litigation adversary is no exception. There are any number of tricks and traps associated with TAR protocols of which counsel and clients should be aware. Of particular importance is whether the litigation at issue even merits the use of TAR, let alone a stipulated use protocol with opposing counsel.
If TAR is a suitable search methodology in discovery, counsel must then decide (among other things) whether to enter into a use agreement with an adversary. As first discussed in a recent post featured on the Relativity Blog, there are three key issues that should initially be explored. How these issues play out could very well determine the success of the TAR protocol.
Tip #1: Understand the Receptiveness of Opposing Counsel
One of the foremost matters to evaluate before entering into a TAR protocol is the receptiveness of opposing counsel to engage in discussions designed to achieve efficient discovery. The ESI search and review process has always been complex; providing opposing counsel with carte blanche participation –as may TAR protocols do – may create tensions given the parties’ adversarial interests in litigation. Along with those tensions is the possibility of satellite litigation with motion practice and corresponding delays, all of which could offset some of the cost and time savings otherwise offered by TAR.
Before approaching opposing counsel with a TAR protocol, a lawyer should first explore the background of litigation adversaries. Will opposing counsel “go along to get along” or will counsel simply make unreasonable demands – known as the “TAR tax” – for the sake of thwarting the use of TAR? Some brief internet research or a couple of phone calls could quickly reveal the answer.
Tip #2: Avoid Improper Disclosures
Another consideration with TAR protocols is that they may contain provisions requiring disclosure of the documents used to train the TAR algorithm. While disclosing those documents—typically referred to as the “seed set”—may be acceptable in certain instances, there are scenarios that militate against disclosure. To be sure, relevant, non-privileged documents used to train the seed set must be produced in discovery. However, there is no obligation under the Federal Rules of Civil Procedure to identify which relevant documents comprise the seed set. This is because a seed set is a selection of documents that may reflect a lawyer’s perceptions of relevance, litigation tactics, or trial strategy. These selections of documents have been protected in analogous circumstances as fact or even opinion work product.
Beyond the issue of work product, a seed set frequently contains nonresponsive documents used to train the TAR algorithm. Besides the fact that such information is outside the permissible scope of discovery, disclosing nonresponsive documents may violate counsel’s ethical duty of confidentiality. Nonresponsive information may contain trade secrets, sensitive financial data, or other proprietary information that should not be disclosed to a litigation adversary, particularly if the adversary is a business competitor. Moreover, depending on the nature of the information, it could be used to add new claims to the present lawsuit or to file a new lawsuit against the producing party. Including provisions in the TAR protocol that protect against such disclosures is an essential step to addressing these issues.
Tip #3: Leave Room for Flexibility
While an executed TAR protocol may look good on paper, sometimes revisions are needed to address deficiencies. Any deviations to the protocol will likely require consent from opposing counsel and court approval. Seeking agreement with opposing counsel may not be possible and a court may not be sympathetic to modifying the protocol. Moreover, as evidenced by Progressive Casualty v. Delaney, implementing unilateral changes to the agreed-upon process may invite the court’s displeasure and adverse consequences to the client. To avoid this scenario, a TAR protocol should provide sufficient flexibility to prevent a litigant from being unreasonably locked into an undesirable process.
Other Considerations
These are only a few of the many considerations that factor into developing a successful TAR protocol. Under the right circumstances, a TAR protocol can yield efficient and cost-effective results in discovery. It can also provide greater certainty that a court will approve TAR-based document productions over the objection of adversaries, as recently demonstrated by the second Dynamo Holdings opinion.
Counsel and clients would be well served to carefully study some of the authoritative resources that industry cognoscenti have developed on TAR protocols. This includes the 2016 Guidelines Regarding the Use of Technology-Assisted Review, published by the Coalition of Technology Resources for Lawyers (CTRL), and the TAR Case Law Primer, published by The Sedona Conference. They should also consider attending the Negotiating TAR Protocols session next month at Relativity Fest where these issues will be discussed in further detail.
