by Constantine Pappas on March 31, 2014
At LTNY 2014, members of the advice@kCura team helped staff the Relativity Showcase throughout the show. Several Relativity users were also on hand to share their experiences with the software.
Paul Laven of Merrill Corporation was one of those users. He shared his story with attendees and the kCura staff and, after the show, sat down with Constantine Pappas to further discuss his experiences with Relativity.
Constantine: What’s been your experience with Analytics for Relativity 8 so far?
Paul: It’s my favorite thing in the world. It’s helped me change the way I’ve been able to pitch the Analytics package to clients and really get them sold on how much it can do to help a case.
Recently, I had a case with 360,000 keyword-responsive documents we needed to produce—after a privilege review—within two weeks. The team wasn’t prepared to hire contract attorneys to get it done, so I ran Analytics to thread all of the emails to help kickstart the process. It took nine days to finish the privilege review with just six attorneys with the help of that threading organization.
The reporting is also great, because we can answer questions about how many threads there are, whether any originating emails are missing from any of those threads, and more. That information is right there in front of you in these reports, whereas without them you’d have no way of knowing that these things were missing. Armed with those kinds of details, we can make sure our collections were complete early on.
Email threading is such an easy thing to deploy on a case, and near-duplicate detection and foreign language identification are also really helpful. What’s great about Relativity is that those items are part of the greater Analytics package, so you get everything else with it, too—including Assisted Review, clustering, categorization, and all the rest.
Do you often mix and match analytics with traditional review tactics?
Absolutely. I constantly try to help people understand that Analytics helps me augment some of the basic review tasks we already perform. You still tag documents, screen for privilege, and so on, but you take out a lot of the clicks, time, and money spent to get those tasks done. It’s still a defensible and repeatable process—it’s just faster. Combining text analytics with searching and other features into a workflow that makes each review as painless as possible helps us do that.
I started using Analytics in 2010, with some clustering on a large defense contract with several million documents. From there, I read a lot about it to learn how it could offer value to more clients. At the time, we were looking at the early stages of computer-assisted review, too, so I got to pilot a couple of those early cases. It was exciting, and I was fortunate to work with partners at clients who were very interested in using the software and seeing what it could do to help them get through their data.
I’ve been using Analytics as much as possible since then. Everyone’s facing the problem of big data now, and Analytics is really about making that less painful for our clients. It’s tough to collect millions of documents and face the daunting task of hiring reviewers to get through them all, so I like putting this technology to work to cut through that.
What kind of impact does Analytics have on your projects?
For us, using text analytics is really about training users and making sure they understand how all of the data interacts, and how these features can help them interact with the data more effectively.
To name one example, I have consistently seen a 100 percent increase in review speeds with email threading turned on. Time and again, if reviewers get through 500 documents a day without it, they’ll get 1000 done with it. That’s an incredible impact, and pays for the technology in and of itself.
A lot of that impact comes from workflow improvements. For example, depending on the data, I can build a workflow that propagates responsive coding on an inclusive email to the rest of the thread. Second-level reviews can adjust the coding decisions based on privilege. That way, you can be sure you’re not missing anything before you produce.
When it comes to productions, we often have cases that produce only inclusive emails, because it means we’re sending out all of the responsive content without having to send a bunch of duplicative text. That’s been pretty successful. In a lot of cases, it’s an obvious payoff—much like global de-duplication. Why produce the same emails over and over again?
On top of those items, language agnosticism in Analytics is a lifesaver. At Merrill, we do a lot of translation work, and we have a great amount of foreign language reviews. To be able to use Relativity for those cases and still get the effective results is fantastic, and it’s not something to be ignored. So many large companies today are international, so it’s priceless to be able to use this on cases with more than just one language in the data set. It’s even great for separate dialects of English, which can be pretty extreme from region to region. You don’t miss terms of art because the system is trained to deeply understand the way people talk.
Why do you think this industry should be ready to embrace text analytics on a wider scale?
This really is not just a wave of the future—it’s the wave of the present. There’s so much data out there that there’s just no way you can read every single thing. Analytics is great because it learns to be incredibly smart when you run it through data, and it helps get through all of the documents quickly. I think people need to get on board with it because there will quickly come a time when we have no choice but to use it. Even now, every case is bigger than the last. Once people start consistently adopting this technology, they’ll see how very few people can gain control of huge amounts of data.
Posted by Constantine Pappas.