Authors Louis Russo and Jacqueline L. Bonneau from Patterson Belknap share this useful article on eDiscovery.

I spoke to my colleague Bradley Koehler about this and we discussed how last month, the New York State Unified Court System adopted new rules and guidelines for eDiscovery.

Here are our takeaways from the changes:

  • First, the rules encourage parties to meet and confer to share information and seek creative opportunities to increase the speed and efficiency of ESI production. This may take the form of conducting phased discovery, which involves prioritising the most relevant Custodians and search terms most likely to yield the highest concentration of responsive material. Additionally, parties who are using search terms to identify ESI for production should review a random sample of documents from search terms returning a comparatively high proportion of unique hits to assess the relevance of those documents and to aid in potential revisions to search terms.
  • Second, the amended rules now include language similar to the proportionality requirements of FRCP Rule 26(b)(1) which involves tailoring requests to weigh the burden and cost associated with conducting preservation and discovery of data sources against the benefit to the case. The rules advise that if a requesting party is willing to foot the cost associated with a discovery exercise that a producing party is claiming is disproportionate, then the court may deny or modify that the producing parties’ burden claim.
  • Finally, while the rules do not prescribe specific guidance on familiarisation with the legal and technical aspects of an efficient and legally defensible process, they do reference the involvement of technical and subject matter experts. Further, parties are encouraged to use technology assisted review (“TAR”) to more efficiently identify ESI for production. The rules state that the use of TAR should not be held to a higher standard than the use of search terms or manual review. There is no opinion on whether TAR 1.0 vs. 2.0 should be used, nor what metrics are most suitable to monitor the accuracy of the TAR process, e.g., statistical sample sizes, recall, precision, F1 measures, elusion tests, etc. However, in practice, parties have been more open to using TAR 2.0 (i.e., CAL – continuation active learning) to assist with prioritizing the review of ESI, rather than using TAR for the purpose of forgoing manual review of documents.