Each week on eDiscovery Leaders Live, I chat with a leader in eDiscovery or related areas. Our guest on November 6 was Ron Best. Ron is an eDiscovery attorney and director of litigation systems with Munger, Tolles & Olson.
Ron and I discussed his current role at Munger Tolles and then moved into a rich discussion about where Ron would like to see eDiscovery capabilities go moving forward. Ron talked about the need for full analytics functionality in eDiscovery platforms. He then laid out his vision for what an “assume everything is possible” eDiscovery system would look like, offering us a play in seven acts.
Recorded live on November 6, 2020 | Transcription below
Note: This content has been edited and condensed for clarity.
Welcome eDiscovery Leaders Live, hosted by ACEDS and sponsored by Reveal. I am George Socha, Senior Vice President of Brand Awareness at Reveal. Each Friday morning at 11 a.m. Eastern, I host an episode of eDiscovery Leaders Live, where I get a chance to chat with luminaries in eDiscovery and related areas.
Past episodes of eDiscovery Leaders Live are available on the Reveal website. Go to reveal.com, select Resources and then select eDiscovery Leaders Live Chat.
This week, joining us as our guest is Ron Best. Ron is an eDiscovery attorney and director of litigation systems with Munger, Tolles & Olson. In those roles, Ron oversees firmwide and case-specific eDiscovery and matter management strategy, implementation, and cost control. Among other things, he serves as a facilitator among firm attorneys, clients, and IT/Practice Support on discovery, fact analysis, and matter management issues involving technology. He has a host of case responsibilities, which I could spend the next hour talking about, but we won’t go into that. He also develops and enhances general firm technology strategy and adoption. This includes needs assessments, vendor qualification, best practices for client extranets, and the like.
Ron holds dual undergraduate bachelor’s degrees in economics and political science from the University of Utah, he received his JD from the UCLA School of Law in 1987, where he completed an externship with Judge Irving Hill of the District Court of Los Angeles. Ron, you and I, we got out of law school at the same time, 1987. Now, one of the things I’m going to ask you about are some of the changes you’ve seen in the intervening years.
Unless I missed something, you started out as a Litigation Associate at Munger, Tolles, from there moved on to the law firm of Hall, Farley, and Obberrecht & Blanton in Boise as a Litigation Associate. Then you switched roles as an Executive Vice President and General Counsel at a company called Fast Check in Salt Lake City, worked as Director of Government sales, Operations and Planning at Micron PC in Boisem and then in 2001 returned to Munger, Tolles where you have been since then. Did I get all of that right?
You have, George. Thanks; good morning to you.
Role at Munger Tolles
Well, thank you and we are pleased to have you here. I’d like you to start out on a very broad topic and we’ll see where you go with that, which is what you do and how you do it when it comes to eDiscovery – you yourself as an eDiscovery attorney, you within the various roles that you fill at the law firm.
Sure. Over the years since I returned to Munger, I’ve worn a number of hats there. That’s actually helped a lot with what I do now. Now, I’m simply back in practice and in essence I help case teams run the information side of their matters. Some of the other hats I’ve worn have been building vendor relationships, building process templates. I put together the firm’s records management policy and process. I was the business person that helped the firm earlier in my time there or my return time there. Specing and rolling out their iManage document management system. I built the practice support teams there in early the 2000s. With all those roles, I got to see a lot of different sides of this thing we call eDiscovery now. What I do now is really help teams gather and use discovery information. That’s a little different than the discovery role when you and I came out of law school. And it has some very significant implications for this practice area we call eDiscovery.
It used to be that discovery was more of a singular event. It was relatively predictable – gather some documents, go through them, turn them over to the case team, and move on to the next discovery project. That’s no longer possible given the volumes of discovery or the volumes of electronic data that now replace paper files that used to be the target of discovery. The amount of time it takes to get through that discovery process pushes the discovery work into the timeline of the case where the attorneys are doing other things. That has fundamentally changed the role of those assisting with discovery; you can no longer simply gather the data and turn it over to the team. You have to go through a data gathering process now that has a great deal more risk associated with it, particularly spoliation claims, and you have to roll information out. At the same time, many of the case teams are doing law and motion work, they’re taking depositions, they’re putting together white papers on specific topics if you’re in an investigation or if you’re trying to inform your client about settlement posture or risk exposure in the case. Not only do we need to manage a much larger amount of information than we used to in the discovery world, but we have to do it on a rolling basis and we have to feed it into these case-specific or matter management functions. I help do that and that’s my primary role, regardless of what I happen to be doing at any moment, which could be preservation or negotiation of ESI protocols or collection or search term testing and analysis or use of analytics. Ultimately the goal of all that is to gather the facts and help the lawyers apply the facts to the law and tell their story.
With that big picture, it sounds like there is an inherent and fundamental challenge. On the one hand, you need to figure out early on so that you can craft your case strategy, what happened and what you want to say about what happened. But we also have more content than ever, more varied forms of content than ever, a lot more to get through and a lot more to figure out. How do you make those two competing sets of factors work together?
It’s a simple concept; it’s much more difficult in practice. Much like the old M*A*S*H series, which was popular years ago, there’s a lot of triage that has to occur. It’s in that triage process where you try… you may have an obligation to gather and get data out to and over the other side, but as you point out, we also need to get to the most important data quickly for our teams. That triage process is about using both experience and new and developing tools to quickly get to what is likely to be the most important information, at least for internal consumption early in a case.
That triage process is about using both experience and new and developing tools to quickly get to what is likely to be the most important information, at least for internal consumption early in a case.
The Need for Full Analytics Functionality
On that tool side, not so much specific tools but types of capabilities, what sort of features, functions, and capabilities are you looking for so that you can make best use of the tools to accomplish what you need?
There’s a large variety of tools out there and available, tool functions, and they’re getting better and better, but ultimately it’s any tool that will allow you to leverage known information and then reach out to the data set and find either validating or contradictory information. We know for example what our clients are telling us about potential exposure or activity in connection with a government investigation. But we need to validate that information or test that information and either confirm or find contradictory evidence. So any tool that allows you to get in quickly. The AI tools, the predictive coding tools, the “find more like this” type of technology, is very useful. There’s been a lot of talk over the years about the clustering technology; that provides some assistance but I don’t find it in practice as useful as some of the better predictive tools. But even things like email threading and near duplicate detection allow you, once you’ve found key documents, to go out and interrogate your data set for anything like it. Where are the drafts? What’s the timing of the drafts for these key documents? Who had them? What kind of social networking occurs around those?
One of the big challenges in our eDiscovery space right now, is that in many, many, systems those types of analytics tools are not available on your entire corpus; they’re available downstream in a review platform. Because of the way data processing and hosting ends up getting priced, we tend to try and collect less and process less and then move a smaller amount into the review world. We don’t get the advantage all the time of that level of analytical interrogation of the data to help us fill out our story. If you can get to a point where you can process and have at least some form of analytics available on the larger corpus, it can really enhance what an eDiscovery practitioner would bring to the case team in terms of the breadth of information surrounding key issues. That’s one of the big institutional challenges I think we face in this industry right now: how do we expand the use of AI and AI related-tools into a broader set of client information.
The Magic Platform: Getting to the Dispositive Story
If I could wave a magic wand and make appear on your screen the ultimate platform for dealing with information and litigation and it could do anything you could imagine – forget about whether anybody actually is capable of delivering this today – what would you like that to look like? What would that magic platform be like?
It’s a great question. Here’s what I think it needs. I think it needs three things that we don’t have fully built in and capable now and I’ll go over those.
#1: In-Place Artificial Intelligence Sitting on Top of Client Data Stores
To begin with, it would sit on top of the client’s data stores. This is a magic wand, so you would get out of the business of needing to extract and process and interrogate subsets of data. We would point this tool to the relevant custodians of the time, whatever happens to be the scope that’s agreed upon, and this tool would interrogate that data in place. It would need to have the full analytics functionality I just talked about; it would need AI, it would need to be able to feed the tool exemplar information, and the tool would then go out and find things related to that.
#2: Seamless Integration of Key Decision Information
That magic wand system is not going to replace the people who need to analyze the information, so that analytics layer needs to be accommodated in the tool. The second thing, in addition to in-place AI, is that the tool needs seamless integration of what I call “matter record keeping”, all the key decisions that we make about “this appears to be a key time frame”, “this appears to be a key set of data from the people involved at the company in this department”, for example. We need to be able to lay that in an, almost like an audit trail, almost like accounting work papers way, in the tool. The basic “who, what, when, where, why” of what we’ve deemed important should be able to be coded right into the tool in some log fashion that’s reportable and can easily be picked up later. Now, most of that type of documentation for most matters occurs in email outside the tools and if you have to go back, it’s very time consuming and very difficult. That’s the second thing I think this magic wand would bring.
#3: Matter Management Capability
The third piece is integration of what we do with the data. So, I call this in the tools, “matter management capability”. Once we begin extracting information out of this great tool – or we don’t even extract it out anymore, see, I’m still caught in our paradigm here. Once we begin identifying information and marking it within this tool in place, we need to use it for the dispute resolution process or whatever it is we’re doing at the time, the investigatory or risk assessment or dispute resolution process. The tool would then have the ability to pivot the information that we’ve identified on site with the client and feed it in the modules, use modules, for a law and motion support, for witness preparation, for fact analysis outlining, which eventually lead to trial issue modules. And the ability for the attorneys to get off of paper outline and mark information in place in these tools, related to the various causes of action or claims and then set up in the tool questions: “We need more information like this.” “Is there anything that contradicts this?” Then, as soon as we receive the information from the other side, we need to be able to lay that right into this same tool, likewise with any third party data that may come in. We need to lay it into the tool and be able to analyze it side by side with the data that our own client has and then begin to draw conclusions about the various aspects of the dispute. So, if you create such a tool, I’m all in.
Okay, well I don’t think I’ll have that ready for you by end-of-day today, but we’ll see what we can do.
The Magic Platform, Part 2
#4: AI Q&A
One thing that I noticed was you talked about questions and having responses to questions. Are you envisioning something, where for example, you could pose a question to the platform and the platform would come back and provide you with an answer as if it were a person talking back to you?
That would be one level of question answering that absolutely would be wonderful. And some of that does exist in some of the AI that’s out there. It hasn’t been broadly applied to the discovery platforms, but that technology is out there and being used in other areas.
#5: AI-Driven Collaboration
But the second question, capability, is more of a collaboration capability on top of the data for the case teams. The case teams have a division of responsibilities. You get a new associate that might be looking at an issue and they think they’ve found something very interesting but they want validation. Imagine if you could, in the tool, have a situation where that question gets attached to the matter outline and the particular fact piece or law piece in the matter outline that’s in the tool. And a notification occurs going out to the person that may be responsible for that particular set of issues. And they can quickly go in and in real time or within an hour or that day see what it is the person saw and then provide responses or instructions.
Again, we do that now but we do it outside the tool. Because it’s outside the tool, we get no audit trail, we get no history. The associate might leave and the fact that a particularly key set of information was identified, may never get answered in our current systems. But if there’s an open question that somebody monitoring the tool can see and the teams involved with analyzing the data can see, then it never gets lost.
When I talk about, in my current matters, facilitating use of the information, that’s what I do now with the current tools. I spend a lot of time funneling our discovery information into makeshift matter management modules that I’ve forced into the discovery tools.
#6: Retain and Reuse
Some of the tool vendors are beginning to come along with the idea that maybe it’s better if we hold on to the discovery information that we’ve processed, facilitated review on, and created a production file that has been sent out. If we hold on to that and facilitate the lawyers using that information, it’s a stickier client experience, both for the lawyers and the ultimate client. Eventually, that same information might be reused in another matter, saving money. And it’s certainly going to be subject to both the law firms and the clients’ records retention obligations. Again, those are areas where Discovery vendors can begin to expand their service offering out of the commoditized processing and review world and begin to provide some real value add. But those tools have to be expanded and drawn out into the “how do we use the data world”, not just “let’s get the data and turn it over”.
I’ll throw one more thought from my end and then we’ll have you close with your thoughts. When I first started practicing law in 1987, one of the partners I worked with said that in an ideal world, when you get a new matter…and we did mostly defense work…the first thing you do is write the jury instructions. Of course you are going to change them as the case goes on, but every time you think, “Should I do this?” or “Should I do that?” or “Should I do something else?”, you should be asking, “Will that get me to those jury instructions?” If it’s not going to do that, why are you even doing it?
Is that concept a part of what you’re talking about here as well?
#7: Dispute Resolution Storytelling – What Trial Modules Do We Need to Be Working On?
Yeah, absolutely. Brad Brian, one of the partners at our firm, who is a trial lawyer, talks about approaching every matter in a similar fashion: “What are the trial modules that we need to be working on?” If you define those early and you filter everything you do in the case in the same manner you’re talking about, by the impact on those trial modules, yes, they’ll change, they’ll morph, one might fall off, you might find a new one, but it ultimately gets you to the goal. The bottom line here is that, as lawyers we need to know the law, we gather the facts, we apply the facts to the law and then we tell a story about that.
The bottom line here is that, as lawyers we need to know the law, we gather the facts, we apply the facts to the law and then we tell a story about that.
That story may be to a judge, it may be to the other side, it may be to our own client to realign the realistic assessment of where the dispute may go. It’s that storytelling that is really the jury instructions and the trial modules.
One of the challenges we face in the discovery area, particularly this eDiscovery area now, is that many of the people that work there are completely disconnected from the dispute resolution storytelling. In many cases that’s a trial, but not always; it could be dispositive motion work, it could be key witness preparation and deposition work that exposes credibility issues that are very important to a case. But the disconnection of gathering of discovery information and how it’s used, means that many people don’t apply the filter you’re talking about. They don’t prioritize to data that will help tell the dispositive story in whatever form that story needs to be told. This recordkeeping that I talk about, this ability to pose questions and interrogate the data, it’s all about getting to those key questions and being able to mark the relevant data inside the tool, so that you can almost automatically build timelines and modules out of that work product, rather than recreating it outside the tool or in another tool.
I think you’re spot on and in a litigation defense mindset, the jury instructions are exactly that, or your trial modules. In an investigation, it may be a little different, but ultimately we’re doing the same thing, we’re getting to the key information. The lawyers that are running the matters and the lawyers in-house that have P&L responsibilities for the matters, they tend to have a pretty good idea of the strengths and weaknesses of the situation. Now, they’re not always going to be 100 percent spot on, but we don’t do a really good job of mining that information early and using that to inform our discovery process to get to the very data you’re talking about. On the bright side, we’re getting better at that, the tools are getting better at that. I think as Discovery practice becomes less about mechanically getting data and turning it over to case teams and more about an iterative process of pulling the important data out of these massive stores that our clients have, we will always, practitioners, get drawn more into that dispute resolution piece and we won’t be pigeonholed in the discovery area.
Well Ron, you have given us a lot to think about and a lot to aspire to. Thank you very much for spending time with us. Ron Best is an eDiscovery attorney and Director of Litigation Systems at Munger, Tolles & Olson, with, as you can tell from our discussion here, a depth and breadth of experience and a range of thought that few can match.
I’m George Socha, this has been eDiscovery Leaders Live, hosted by ACEDS and sponsored by Reveal. Thanks for joining us today, please join us again next week when we talk with Aaron Bath, who is National Director of Litigation Management for Balfour Beatty US, an an industry-leading provider in the US of general contracting, at-risk construction management and design-build services for public and private sector clients across the nation.