Loudoun judge approves predictive coding results

In the first case to approve the use of predictive coding over a party’s objection, a Loudoun Circuit Court judge has signed off on the defense’s computer-assisted review results in a document-heavy case.

Computer-assisted review made headlines over the past year after U.S. Magistrate Judge Andrew J. Peck of the Southern District of New York issued a landmark decision in support of computer-assisted coding and review for electronic discovery.

The Da Silva Moore v. Publicis Groupe decision recognized predictive coding as an acceptable way to search for relevant electronically stored information “in appropriate cases.”

Litigants and courts across the country took notice and the use of CAR has been on the rise ever since.

One of those cases involved the collapse of a hangar in a February 2010 snowstorm.

Pittsburgh lawyer Thomas C. Gricks III represented the defendant in the suit, filed in Loudoun Circuit Court.

“We had a boatload of data,” he said, estimating around 2 million documents. “We knew that by and large, a lot of that stuff was going to be unnecessary to the case and unrelated.”

So Gricks asked the court’s permission to use CAR in the suit.

Last April, Loudoun Circuit Judge James H. Chamblin issued an order in Global Aerospace Inc. v. Landow Aviation (VLW 012-8-207), allowing the use of computer-assisted review over the objection of plaintiff’s counsel.

The technology was “the most efficient and cost-effective way to find the relevant stuff,” Gricks said.

The defense then faced the task of employing predictive coding and producing defensible results.

Input for the predictive coding set was reduced from about 2 million down to 1.3 million documents. Of those, 5,000 random documents were coded as the basis for the remaining electronically stored information, leading to a reduction from 1.3 million down to roughly 173,000 documents that were ultimately produced.

“Being able to cull out 87 percent of the documents in the case without having to review them was a huge cost-saver for the client,” said Karl Schieneman, president of Review Less in Pittsburgh and a predictive coding consultant on the case.

Gricks called the result “very positive,” especially with a validation rate of 81 percent.

“That is significantly better than what you would typically expect with human review, which studies show [has a validation rate] just south of 60 percent,” he said.

Computer-assisted review allows attorneys to “cut down the universe [of ESI] and do it better, faster and more cheaply than a human review process,” Gricks said.

Cautioning that he was calculating a “back of envelope” estimate, he said the technology saved his client between $500,000 and $800,000.

“As predictive coding becomes more and more accepted by the courts, clients are going to anticipate the savings in time and money [the technology] will provide,” Gricks predicted.