Electronic Discovery & Computer-Assisted Review

Posted By: Timothy T. Emmart

RND_0906_0035_v3cYou work for a large firm that specializes in mergers and acquisitions, or perhaps class action suits. Much of the work that you do requires you to sift through mounds of documents, spreadsheets, etc… Increasingly, in the information age, you must also swim through a sea of electronically stored information (ESI). Wouldn’t it be nice if there were a way to efficiently mine that ESI? As it turns out, the answer to that question is here.

What is E-Discovery?

Electronic discovery (e-discovery) in the legal field is an area to keep your eye on. E-discovery is the process of seeking, locating, securing, and searching electronic data with the intention of using the data as evidence. Additionally, court-ordered hacking under the guise of obtaining evidence is also considered a type of e-discovery.

The process of discovery is an expensive legal process. In the analog world, discovery would include pouring over countless documents and analyzing them manually. This work is extremely time-consuming and the legal fees can add up very quickly. E-discovery presents similar problems, but perhaps creates an even larger problem. Legal fees can mount in e-discovery in much of the same way as in traditional analog discovery, but the process of e-discovery may require the hiring of a forensic computer expert, which could serve to make e-discovery even more expensive than traditional discovery.

What is Computer-Assisted Review and How Does it Work?

Predictive coding or computer-assisted review is likely to change the way e-discovery is conducted for years to come. Kerschberg describes predictive coding as “the electronic coding, organization, and prioritization of entire sets of [omitted] [ESI] according to their relation to discovery responsiveness, privilege, and designated issues before and during the legal discovery process.” In general, ESI is information that is created, used, and/or stored in digitally, and requires a computer or other similarly purposed device for access. More, ESI may come in many shapes and sizes, including: textual communications, such as emails, texts, chats, and instant messages; audio/visual files; office documents, such as spreadsheets and word processing; Internet activity, such as browsing history; activity on social networking sites; and computer programming information.

A 2015 case out of the U.S. District Court for the Southern District of New York, Rio Tinto PLC v. Vale S.A ., is thought of as the first court to allow computer-assisted review in the United States. In that case, the parties stipulated to a certain set of governing protocols and the court approved them. However, the court was careful to point out that it was not adopting the agreed protocols as the rule going forward in all cases, but it did note that it is “inappropriate to hold [computer-assisted review] to a higher standard than keywords or manual review. Doing so discourages parties from using [computer-assisted review] for fear of spending more in motion practice than the savings from using [computer-assisted review].” While the Rio court did not adopt a specific computer-assisted review protocol, the court did recognize the need for this type of review when the court approved its use. Moving forward computer-assisted review is sure to become increasingly used in e-discovery.

Computer-assisted review works by amalgamating three elements: 1) a domain expert, 2) an analytics engine, and 3) a method for validation. First, on the front-end, the domain expert creates the framework of the review process and serves as its orchestrator. Second, the analytics engine is composed of indexing and retrieval technologies, which allow documents to be sorted according to domain expert command. Third, the validation component enables the lawyer to provide an audit trail for the efficacy of the computer-assisted review system. These elements work synergistically to process e-discovery at a lower cost and more efficiently than manual review.

Benefits of Computer-Assisted Review

The benefits of computer-assisted review include cost savings and efficiency. A slew of technology companies are jockeying for position in this relatively new space, including Kroll, Symantec, Hewlett-Packard, and IBM to name a few. An article published on americanbar.org declared that there are two main reasons to embrace computer-assisted review. The articled stated that “empirical data scientifically establishes that [computer-assisted review] equals or exceeds human manual review in search and production reliability [and computer-assisted review] reduces the expense of document production, especially in cases involving many gigabytes and/or terabytes of electronically stored information.” Thus, proponents argue the cost savings and review efficiency.

Problems with Computer-Assisted Review

Computer-assisted review is not without practical concerns. One problem is that the relevancy determination is not straightforward, because its determination is subjective, which creates a problem in being able to design a computer assisted review process to account for the subjectivity. Another problem is the risk of over production of privileged documents. This is problematic because some privileged documents may be difficult to identify; therefore, some of the privileged documents are at risk of falling through the cracks. Some argue that the quality of current computer-assisted review software is not up to standard, the gold standard being that which is on par with a properly conducted manual review. Others argue that Daubert, otherwise known as Federal Rule of Evidence 702, should apply to computer-assisted review of e-discovery. As with any new technology there will be a debate over how the technology will comport with existing process. The legal field is no different, and many times the legal field is late to adjust to new advancements, be they technological, medical, social, etc… Still, there is a clear need for computer-assisted review in e-discovery at least in complex cases, such as in class-action disputes.


As the legal arena continues to both approve the use of computer-assisted review and deal with its related issues it will be interesting to watch the process unfold. Perhaps best practice protocol software will emerge. Surely, computer software creators will continue to strive to find the golden ticket, the software that solves the e-discovery riddle once and for all. Until that time, keep an eye on it.


Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s