fbpx
Home / Law / Computer-assisted review aids discovery

Computer-assisted review aids discovery

The future is here.

A recent decision from the Southern District of New York may ease the reluctance of attorneys to use computer-assisted coding in the discovery process.

U.S. Magistrate Judge Andrew J. Peck, of the U.S. District Court for the Southern District of New York, has issued what is likely the first ruling on its use.

In Da Silva Moore v. Publicis Groupe (No. 11 Civ. 1279 (ALC) (AJP) (SDNY Feb 24), Judge Peck approved the use of computer-assisted review in a gender discrimination case brought by five female employees against an advertising conglomerate.

The employees alleged gender discrimination under Title VII, pregnancy discrimination under Title VII and the Family and Medical Leave Act, as well as violations of the Equal Pay and Fair Labor Standards acts.

During discovery, both parties had agreed to use computer-assisted review, but the plaintiffs objected to the methods used by MSL Group, the U.S. public relations subsidiary of Publicis Groupe.

Computer-assisted review, unlike manual review of documents, involves the use of computer technology with sophisticated search methods to help attorneys quickly find information relevant to the matters they are litigating. Predictive coding is the electronic coding, organization and prioritizing of sets of electronically stored information.

The plaintiffs opposed MSL’s use of predictive coding to try to cull down the approximately three million electronic documents from the agreed-upon custodians, claiming MSL had oversimplified its use. They also questioned its reliability.

Judge Peck ruled their concerns about reliability were premature and that, in this case, the use of predictive coding was appropriate based on the fact the parties had agreed to computer-assisted review. He also cited the volume of the documents (more than three million) to be reviewed, the superiority of computer-assisted review to linear manual review or keyword searches, the need for cost effectiveness and proportionality under Rule 26(b)(2)(C) and the transparent process proposed by MSL.

“This opinion appears to be the first in which a court has approved the use of computer-assisted review,” he wrote. “That does not mean computer-assisted review must be used in all cases, or that the exact ESI [electronically stored information] protocol approved here will be appropriate in all future cases that utilize computer-assisted review.”

Judge Peck also pointed out he was careful not to mention the names of party vendors in the body of the opinion, so as not to endorse any one.

“What the bar should take away from this opinion, is that computer-assisted review is an available tool and should be seriously considered for use in large-data-volume cases where it may save the producing party (or both parties) significant amounts of legal fees in document review,” Judge Peck wrote. “Counsel no longer have to worry about being the ‘first’ or ‘guinea pig’ for judicial acceptance of computer-assisted review.”

Judge Peck is the author of an October article in Law Technology News called “Search Forward: Will manual document review and keyword searches be replaced by computer-assisted coding?” He quoted from the article in his 26-page decision, dated.

“Judge Peck is a proponent of using technology in the legal process and discovery,” said Peter Coons, co-owner and senior vice president of D4 LLC, a national company providing litigation support and e-discovery services to law firms and corporate law departments.

Coons, who works out of the Rochester office, said the significance of the decision is using computer-assisted review technology to identify documents that had once been reviewed initially by a war room full of attorneys.

“The technology and the law are coming together,” he said. “Up until this point, there hasn’t been an opinion that discusses predictive coding.”

D4 uses predictive coding technology to help its clients. Coons said it has been available for a long time, waiting to gain acceptance.

He compared it to a tool to build a house. He said someone has a hammer and D4 represents the carpenters that help build the house, noting there are still other components necessary such as building permits and painters.

Coons estimates the use of computer-assisted review could save companies millions of dollars, costing about 1/10th as much as a linear review.

He said it is much different than a traditional document-by-document review. A lead attorney, someone who knows the case, will “train” the system, specifying small sets of data, based on what documents are to be reviewed to selectively pick out the information highly relevant to the case. Results may be further refined to narrow the scope.

“It augments the discovery process,” Coons said. “It doesn’t replace human eyes. We’re not at that point. It does utilize technology to winnow down that volume of data. It’s not just producing relevant documents. It’s finding the highly relevant documents that are really going to be the foundation for your case — the proverbial smoking gun.”

Alan M. Winchester, a partner in the law firm Harris Beach PLLC, said the Da Silva case is not all that significant in and of itself, but that there is a tremendous amount of money spent in electronic discovery. He has seen estimates in the billions of dollars.

Winchester works out of the New York City office and is leader of the firm’s Electronic Information Counseling and Management Team.

“The case, I think is a great first step in that area, but it’s just that, a first step,” he said, noting a lot of attorneys might have concerns with the methodology in Da Silva.

Winchester said the seed documents may produce documents a firm does not want to turn over in discovery such as employment records.

He cites a federal case out of Chicago, Clean Product LLC etc. v. Packaging Corporation of America, which he said is the opposite of Da Silva.

“I guess in that case, the plaintiffs didn’t like the documents they were getting,” Winchester said. “They said the defendant used keywords instead of coding.”

He said the parties should have “a meaningful meeting” to discuss what key terms best describe the subject documentation, as well as agree on what terms to exclude.

He compared it to looking at banker’s boxes in a warehouse where one might look at a label and determine the contents are not what they are seeking until they come across the box or boxes they do care about.

“It’s really the same think with a computer now,” Winchester said. “The computer’s going to make these banker’s boxes. Even in a warehouse, you come to a box labeled junk. Why look at every single document?”

Coons and Winchester agree perhaps the greatest challenge to the use of computer-assisted review is a lack of understanding and the need to be knowledgeable and involved in the process.

Coons said the main challenge is that its use is not mainstream, but that cases like Da Silva are helping remove apprehension.

“I’m a believer,” Winchester said. “I don’t see why there would be any concerns. It’s going to give a better product for less money and faster than this linear review. It’s like any tool. If you know how to use it, it can bring great results.”

Winchester said there are other benefits like quality checking and increased accuracy. The software also aids in meeting tight deadlines.

“Accuracy is going to be greater than a human review because humans make a ridiculously large number of mistakes.

“It is going to be a lot less expensive in terms of hours attorneys need to be looking at stuff because their time is going to be focused on looking at the documents that matter,” Winchester said. “Now, huge amounts of analysis will be done up front.”

He said there is still a role for document review specialists; just maybe not as many will be needed. He estimates firms can save time and money by sheer volume; needing to look at only 20 percent of the documents involved because the other 80 percent will be weeded out as non-responsive.

“There’s probably a better value the more documents you have, but I don’t think it doesn’t have value in smaller cases,” he said. “The benefit grows with the more documents.

“It’s uncanny the way it groups similar documents. I can see why there is a wizardly feeling to it. How could it possibly realize that all of these documents are alike? I think people just need to know that while it may be new to e-discovery, it’s not new to the world,” Winchester said.

Leave a Reply

Your email address will not be published. Required fields are marked *

*