Your single source for new lessons on legal technology, e-discovery, and the people innovating behind the scenes.

Beyond Rio Tinto: Real-world Proof Points for Analytics

Andrea Beckman

Earlier this month, we posted a two-part series on our exciting—albeit rocky at first—journey with analytics in e-discovery, as well as some takeaways from Judge Peck’s recent Rio Tinto opinion. As we’ve mentioned around this topic, there’s been a lot of growth in analytics usage over the last couple of years.

But how has all this translated to tangible benefits? We’ll break it down into our three goals for analytics in e-discovery: making e-discovery faster, easier, and more insightful. Check out a few real-world stories below, and visit our Customer Wins page for more.


Law firm Thompson & Knight recently approached Inventus, an e-discovery service provider, for support on an internal investigation that required an immediate start to review—and by “immediate” we mean they collected data on a Friday and needed prioritization done by Monday.

Though initially wary of the timeline, the Inventus team set up a computer-assisted review project to prioritize the 86,000 documents a keyword search returned from the initial 1.8 million-record set. After only one training and two QC rounds, the team achieved an overturn rate of just 9 percent.

With this analytics workflow, the case team managed to review just 2,100 documents and immediately prioritize a responsive set of about 12,000 records—all in a single weekend.

“During this project, Assisted Review provided immense value in an incredibly short period of time.”

– Alex Jacobs, senior discovery consultant at Inventus

Ease of Use

Assisting a client facing a government subpoena, law firm Troutman Sanders faced a review of 742,000 documents—59 percent of which were emails spanning 60 different custodians.

The team planned to use conceptual analytics to churn through the data, but knowing that 441,000+ of their documents were emails made them confident that email threading—a feature that automatically groups emails by conversation—could lead to an easy win early on in the project.

Email threading immediately identified just 293,000 of those emails as inclusive—meaning they contained unique content, or all the content in an entire conversation. By focusing on only these records, the case team reduced the reviewable set by 34 percent and saved their client approximately $233,000.

“Aside from helping us save time and money, the whole process took only a couple of hours to set up—which is just another reason it’s so enticing. It’s easy to use and a major cost savings.”

– Chris Haley, director of litigation technology at Troutman Sanders eMerge


In the midst of an insurance matter, law firm Stradley Ronon Stevens & Young received a production of around 7,900 documents one Friday morning. The partner on the case wanted to see all of the key documents by Monday.

An associate took ownership of the review, despite the fact that he was new to Relativity. He contacted Brendan Curran, the firm’s litigation support manager, to get started. Brendan quickly set him up with a workspace employing clustering—a feature that automatically groups conceptually related documents.

While Bradley walked him through the technology, the associate dove right in. In the end, he got back a lot of the time he’d expected to lose: he spotted a single cluster of 450 relevant documents almost immediately, and decided within minutes that these were the only records he needed. He completed the review in just a few hours—getting an in-depth look at the technology and its benefits, as well as at the data itself.

“This associate—someone who had never used Relativity before—was able to quickly learn how the system works and then find exactly what he needed.”

– Brendan Curran, litigation support manager at Stradley Ronan

Is Analytics Coming Together?

It took several years before case teams began leveraging text analytics in earnest. But today, case teams aren’t just using it more often—they’re asking for more sophisticated analytics tools to visualize, dig deeper, and gain insights from their data. So even if e-discovery software has come a long way in making text analytics fast, easy to use, and insightful—we’ve got plenty of room to keep improving.

If you’re looking for information about text analytics and how the technology can support your projects, check out this e-book. It’s a brand new, plain-language introduction to these features and what they can do for your team.