Caragh Landry, president of litigation services, TCDI

While analytics and Big Data have been popular in the e-discovery technology realm for a few years now, that popularity doesn't seem to have carried over quite as far into e-discovery practice as many would like.

Caragh Landry, e-discovery group TCDI's new president of litigation services, is hoping to change that. A former global head of managed document review at e-discovery group Integreon, Landry has seen e-discovery transition from scanning and coding boxes of documents into a far more digital process, but one that still seems to lack a fair amount of analytics-based oversight.

Landry spoke with LTN about her transition to work at TCDI and the shifts she's seen in the industry in recent years:

Where did you grow up? New York, Connecticut and Illinois, mostly Connecticut.

How did you get into the e-discovery world? I was working as a corporate trainer for a large hotel brand in San Francisco and got burned out on customer service. I answered an ad for a QC Supervisor at DRI in SF (which changed its name in 2000 to CaseCentral), and it turned out to be one of the earlier online e-discovery companies.

You've worked now at a few different e-discovery shops. What factors make for a successful working relationship with clients? For me, it's respect. The most successful relationships I've had with clients are those where we all respect each other's experience, come to the table looking for the best solution and are really interested in solving the problem together. This viewpoint has also proven successful in those situations where I've worked with the clients' other partners (i.e., competitors of mine) to find a solution. I've found that, if we put aside our traditional ideas of competition and value the end goal (what's best for the client), that we can openly work with each other to get to an impactful solution. And that can't be done without respect.

How have technology advances changed e-discovery processes throughout your career?It's been interesting for me over the years to see the adoption, or really lack of adoption, of all the various analytics tools. Unfortunately, I don't think those tools have yet to realize their potential.

What kinds of technology or process shifts do you anticipate in coming years? I hope to see significant changes in the document review process itself. While review is now done mostly online, digitally, the overall process has not changed too much, even with the advent of analytics. As the purpose of discovery is to find relevant material—our intent at TCDI is to work with legal teams to approach document review differently. We'd like to see document review regarded more as an investigation than as a linear process.

Continuous active learning (CAL), in our minds, is what is really going to shake up the document review process. With this in mind, we see the business of document review changing to support that new approach, too—smaller, more expert teams are what will be required. The days of 100- to 200-person review teams will go away, and we also expect to see the type of resources (i.e., nonlawyers) doing some of the document review change as well.

Artificial intelligence: overhyped or worth the hype? Both. I think unsupervised machine learning is nothing more than grouping similar documents together, and people are freaking out and not using something that they should actually be insisting on. Why not group similar documents for better viewing, consistency, accuracy and speed? I think supervised machine learning, as in the kind we see in CAL, is way undervalued.

I see our AI tool as the paralegal who has been working on the case for 10 years—she's seen and read every document, and she knows where everything is. And I use our AI tool (we call her Tara) the same way I would use that paralegal. I tell Tara what I saw, what I liked and what I want to see more of and then she goes out and finds me more documents to look at, that are like the ones I liked. She know what to provision for me because I told her what I like, and she's knows where everything is. So, for me, I think AI in legal is both overhyped, made to seem scary, and totally worth the hype. (You need to meet Tara. She's awesome.)

What are your favorite and least favorite e-discovery jargon/buzzwords? Speaking of AI, it is my least favorite buzzword—I think it's too broad of a term, is often misused, and people are scared of it. They think SkyNet, they think unemployment, and they think confusing.

My favorite buzzword is “e-disco.” I remember people calling e-discovery “e-disco” about 20 years ago, but I didn't think it stuck until I recently found a Reddit subpost called “eDisco.” Awesome. Come for the cats; stay for the eDisco.

One thing you're hoping will change about the industry in the next year? Analytics. I know, I know, it's my same answer every year. But 2018, I feel it, this is going to be the year where people finally get that analytics are a game-changer and worth every penny in terms of document review. I am amazed that I still meet clients who don't even review documents by thread complete or group near-dupes together—they still review by custodian and sort by date. Why someone would want different people reviewing various pieces of an email conversation or have different reviewers viewing slightly different versions of contracts and presentations perplexes me.