In Part I and Part II of this series, we discussed the key takeaways from Principles 1-3 and 4-6, respectively, of The Sedona Conference WG1’s “Commentary on Defense of Process: Principles and Guidelines for Developing and Implementing a Sound E-Discovery Process” (available for download here). The Commentary seeks to address what parties can do to avoid, or at the least prepare for, challenges to an eDiscovery process they apply in a given matter and how courts should address discovery disputes.
The following are the key takeaways from Principles 7-9 of the Commentary.
Principle 7. A reasonable e-discovery process may use search terms and other culling methods to remove ESI that is duplicative, cumulative, or not reasonably likely to contain information within the scope of discovery.
Comment 7.a. Eliminating ESI that Is Cumulative or Duplicative
This principle is focused on striking the right balance between producing significant, unique, relevant ESI and holding back cumulative or duplicative ESI. To help achieve this goal, the Commentary suggests the use of various techniques, including de-duplication and email threading.
Comment 7.b. Eliminating ESI that Is Outside the Scope of Discovery
The Commentary proposes that in most cases, a culling process that removes a significant volume of irrelevant or duplicative information is reasonable, even if it also removes some relevant information. Obviously, it is best to reach an agreement on the culling strategy with the requesting party before proceeding.
Comment 7.c. The Use of Search Terms to Cull ESI
Search terms are a defensible technique for limiting the document population for review. However, developing search terms requires eDiscovery expertise, as well as knowledge of the particular case. Courts have emphasized the importance of some of the following processes when developing search terms:
- Consultation with persons knowledgeable about the relevant subject matter (g., relevant terminology, abbreviations, terms of arts);
- Consultation with persons with appropriate technical expertise to ensure the proper use of syntax and technological tools;
- Use of an iterative term-selection process where two or more sets of searches are run to help refine existing terms until a reasonable result is achieved;
- Documentation of the search process; and
- Disclosure to or cooperation with the requesting party.
While disclosure and cooperation are expected and can aid defensibility, Commentary suggests the responding party should not permit the requesting party to dictate “unreasonable” search terms that would create an undue burden relative to the likelihood of identifying additional responsive ESI.
Comment 7.d. Other ESI Culling Techniques
The Commentary suggests the use of technology-assisted review (or TAR) in lieu of, or in addition to, search terms, as well as the application of date-range and file-type filters and identification of “near duplicates” or conceptually similar documents. Since these methods can be expensive and time-consuming, it is advisable to seek an agreement with the requesting party regarding the specifics of the process.
Principle 8. A review process can be reasonable even if it does not include manual review of all potentially responsive ESI.
Recent studies suggest that TAR review yields more accurate results than exhaustive manual review. Other benefits of TAR review include relatively low costs when compared to manual review, lower efforts, and quick results.
The eDiscovery scholar, Magistrate Judge Andrew J. Peck of the Southern District of New York reached the following conclusion in Da Silva Moore v. Publicis Groupe, 287 F.R.D. 182, 189 (S.D.N.Y. 2012):
Linear manual review is simply too expensive where, as here, there are over three million emails to review. Moreover, while some lawyers still consider manual review to be the “gold standard,” that is a myth, as statistics clearly show that computerized searches are at least as accurate, if not more so, than manual review.
The Commentary suggests that at this time there is both sufficient empirical evidence and sufficient case law to conclude that TAR is a “reasonable” way to search for responsive documents.
It is important for parties to consider how their jurisdiction views the use of TAR and whether their particular case would benefit from it. As with the rest of the eDiscovery process, to the extent possible, it is advisable to attempt to reach an agreement regarding the parameters of that use with the requesting party.
Principle 9. Technology-assisted review should be held to the same standard of reasonableness as any other e-discovery process.
Comment 9.a. The Application by Some Courts of Differing Standards
The Commentary calls upon courts to apply the same level scrutiny to the use of TAR as they apply to the use of search terms to cull ESI. To date, the use of TAR has tended to be subjected to higher standard and oversight compared to more traditional methods. This can lead parties to avoid the use of TAR in favor of traditional, more time-consuming and labor-intensive, methods that may not produce better results.
As such, it is important for parties to be aware of how their jurisdiction approaches the use of TAR and whether it will impose more rigorous requirements for transparency and validation. While TAR has many benefits, when involved in a case in an “unfriendly” jurisdiction, a party may choose to undertake the expense of manual review to avoid subjecting itself to increased scrutiny.
Comment 9.b. Common Standards Should Apply to All Review Methods
Carrying forward the comment above, the Commentary argues for equal treatment of manual review and TAR methods. As an illustration to this point, the Commentary suggests that when a party has applied a TAR process to review hundreds of thousands of potentially responsive documents and the requesting party subsequently demands a level of validation that would not have been demanded if the responding party had chosen a manual review process, the responding party should have the right to reject these demands because they would hold one review process to a higher standard that the other.
As technology evolves and volumes of ESI increase with each passing year, the eventual wide-spread acceptance of TAR appears inevitable. In the meantime, parties should consider whether their particular matter is appropriate for the application of TAR, the specifics of such application given the circumstances, and the amount of effort and level of disclosure that could be required to reach an agreement on the use of TAR in a case, as well as the level of risk-taking involved in a unilateral decision to use TAR.