Discussing developments in technological innovation in legal practice
Paul-Erik Veel is a Partner at Lenczner Slaght LLP. On January 14, Veel sat down with Ultra Vires to discuss his experience leveraging data and technology with his work at the law firm. He shared his thoughts on how to foster a culture of technological innovation in legal practice, explained some of the limitations of legal technology, and reflected on where he sees technology in legal practice developing in the next decade.
This interview has been edited for brevity and clarity.
Ultra Vires (UV): There has been an increased interest in incorporating technological solutions to legal practice. A few years ago, you started the Data-Driven Decisions program at Lenczner Slaght LLP. Tell me more about that.
Paul-Erik Veel (PEV): Data-Driven Decisions as a program is about using empirical and quantitative insights to inform the practice of lawyering.
I was working with a client in a case where we were going to a jury trial. It was a medical malpractice trial and I was giving the doctor some advice about how he should do such or such. I don’t remember exactly what it was, but I was giving him some very conventional advice about the way we do things and what he should do. And he, being a doctor and trained in a very empirical background, said, “Okay, great, thanks very much. And just to confirm, of course, there [are] studies that validate that that’s the right strategy, and will increase my chances of success?” to which the answer was, well, no, that’s not actually how being a lawyer works. We don’t have empirical data. And the vast majority of academic work actually doesn’t focus on those types of questions. That was the honest answer. But it was an unsatisfying answer to him. And it was unsatisfying to me as well.
Since then, with the support of my firm and lots of colleagues, we’ve tried to incrementally change that by collecting data about things that are relevant to our practice and then leveraging that data to give our clients better, and more empirically-grounded advice. Clients will often ask their lawyers “what are my chances of winning this case?” And in the olden days, a lawyer’s answer was, “well, I can’t really tell you, but you’ve got a decent shot.” What is a business person—what is a rational client—supposed to do with that information? It’s very difficult to understand what that means. I’m seeing more and more clients that want to know: “Okay, what does that mean? Do I have a 70% chance of winning? Do I have a 50%? Do I have a 30% chance? Put some meat on the bones of your estimate.” That’s not something lawyers were comfortable doing historically, in part because it’s very hard to give those probability assessments without a broad range of data to inform them. So at its core, Data-Driven Decisions is about using and leveraging that empirical data to advise and advocate for our clients better.
UV: One of these applications at your firm is the Supreme Court Leave program. How does the data process work in the Leave program and what has been your experience with it in terms of successes?
PEV: From my perspective, our Supreme Court Leave program is actually one of the best examples of where a quantitative empirical analysis can be helpful because, as you know, the Supreme Court [does not] tell you why they grant leave in cases or not. Instead, every Thursday morning, they will release decisions just saying that leave applications were granted or dismissed, and you have no idea why they’re doing what they’re doing.
I suspected that there were certain regular patterns that we could identify that would help us predict the likelihood of particular cases getting leave to the Supreme Court. When we started collecting the data, we had a whole series of hypotheses as to what might be relevant and what might not. We didn’t know what the data would show, but we hoped that we would find some fairly strong relationships. Lo and behold, once we actually collected several years’ worth of data and analyzed it, it did.
There is a whole laundry list of factors that we can say not only impact your chances of getting leave, but we actually have an estimate as to how they change your odds of getting leave. And that all feeds into a model that gives predictions. Getting leave to the Supreme Court is always unlikely; in past years, it’s been in the six to eight percent range of cases that get leave every year. So, our model does not tell you with certainty whether you’re going to get leave; it doesn’t give you a yes/no prediction. What it tries to do, and what I see as its utility, is that it allows you to stratify your case by how good a shot you have. Is this a case where I feel like the Court of Appeal did me wrong, but the model says I’ve got a one percent chance, so it’s not worth spending any more time or effort on it? Or is this a case with a 30 or 40 percent chance where, maybe it’ll happen, maybe it won’t, but it’s worth spending the effort to pursue it to the next step?
UV: Every leap in technology comes with many challenges and hidden costs. What are some of the biggest challenges that you have experienced with the program and what did you do to ameliorate them?
PEV: I think the challenge in Data-Driven Decisions is that there’s always a change management element. With a lot of innovations, it’s hard to get people to adjust to a new way of thinking about things. The legal industry is inherently conservative in so many ways. There have been a lot of people who have remarked that a lawyer practicing 100 years ago, if you plop them down into a legal practice today, they would be much more comfortable doing their job than a doctor or a dentist or an accountant or one of any other panoply of professionals.
When it comes to numbers and data, regrettably, a lot of lawyers are allergic to using them. And that comes from a lot of different places, including the fact that for a lot of lawyers, their background before law school was not in a field that used data or math or statistics to any meaningful extent. I think there’s a lot of hesitancy, and oftentimes—not so much within my firm—but I can tell you other lawyers sometimes have had this reaction as well, “And what’s your model going to be able to do? What can aggregate data tell me?” My response to that is always, “judicial analytics and empirical legal analysis is not going to fundamentally change the way you practice. But it’s an additional tool that can improve your practice and allow you to do your work better, faster, cheaper. It’s a second opinion. It’s a first pass. It’s a gut check. It’s a starting point and an input for you to formulate your advice and act for your client.”
To come back around to the question you asked, to get people to buy into a more data-driven mindset, you have to, on the one hand, show them concrete, tangible benefits, but also make sure that you’re not over-promising the capacity and that they understand what it can do and what it can’t. Once you have the right mindset, it’s an immensely powerful tool that can complement all of the conventional lawyering that we do.
UV: The legal field is often regarded as one of the fields that is least vulnerable to automation compared to other occupations. To what extent do you view the development of technology as a means to either replace or supplement a lawyer’s work going forward in the next decade?
PEV: I would be shocked if there is a meaningful change in what lawyers are doing today, compared to what they will be doing a decade from now. There will of course be a change. There will be additional tasks that will be automated. But a lot of the lowest hanging fruit has already been picked.
I’ve been at Lenczner Slaght for 11 years. When I started practicing 11 years ago, document review was still a relatively manual exercise. You would do keyword searches to try and identify relevant documents. If you had a set of one million documents to look through, the common way you would deal with that would be to hire a review team, write a coding manual, and give them instructions as to how to review documents. It was inefficient and expensive. Nowadays, that still happens sometimes. But, most of the time, you do some kind of technology-assisted review where you will have a machine learning tool that uses natural language processing to do a first pass analysis of the relevance of a particular document. That’s already happening. And there are still all kinds of advances being made every year, but the low hanging fruit has been picked.
I think we’re going to see incremental innovations in legal research, document assembly, and document review over the next decade. But I think it’s going to take longer than a decade to see fundamentally transformational change to the way that lawyers lawyer.
UV: The pursuit of technology may not be a priority from a day-to-day perspective in legal practice today. What has been your experience on how law firms should organize to foster a better environment for the development of technological innovation in their practice?
PEV: Innovation is one of those things that costs money and yields no good outcomes in the short term. In a law firm that is taking a short-sighted perspective and is not taking the steps necessary to incentivize innovation, management isn’t going to send the signals necessary to say to lawyers: “We value you working on innovative activities.”
That being said, there are a handful of law firms out there—and I count myself lucky to be at one of them—that understand that if you’re trying to serve clients at a high level, you must continually be innovating to try and improve the quality or lower the costs of your service delivery. What does that mean in concrete terms? It means firms can do all kinds of things, but the number one part of innovation is just a culture at the top that says, “if you have an idea that seems kind of crazy, pursue it, and if it doesn’t pan out, and if you fail, and if you put in 100 hours of your time, and it fails, that’s okay because if you don’t do that, we’re not going to generate innovation.”
Different firms do it in different ways. I’ve seen some reports of US firms that have billable hours targets that allow associates to count a particular number of hours towards sort of innovation-focused activities. Billable hours targets are in and of themselves controversial, so I’m not going to weigh in on the rightness or wrongness of that. Certainly, those kinds of signals show that innovative activity is encouraged and failure is okay, that the firm is willing to spend upfront money and think about different ways of doing things. That’s what matters more than anything. There’s no sort of magic to innovation, other than having a group of smart people together, and being given the support to follow their seemingly crazy ideas, wherever that may lead.
UV: Is there anything else that we have forgotten or that you would like to add?
PEV: The only thing I would add is this. There really are a lot of different ways to be a lawyer. Even the largest of firms and the most white-shoe Bay Street firms are recognizing that there are new ways to do things that are different from how we’ve done things in the past. And that’s great! So what I would say to every U of T Law student, and every law student in general, is as you’re figuring out your path through all of this, the law is complicated and there will be so much that you can learn from the people more senior to you. But anytime you have a good idea, put your hand up, and just take the initiative to follow it. Ultimately it will make you a better lawyer and, in the long run, that’s the way that the industry as a whole is going to get better. So, when you have a crazy idea, follow it.
