#70 Data, AI, and the New Legal Ops Playbook

Hello everybody, and welcome to Pearls on Gloves Off. I'm your host, Mary o', Carroll, chief operating officer at Goodwin. Today's conversation is one I've been really looking forward to. I'm joined by my friend Aaron Brummagem, head of legal operations at Snowflake, a company that's right at the center of the AI data revolution. But Aaron's story and his role goes far beyond that. He's a 20 year tech veteran, a startup founder, one of the earliest Yahoo employees, and someone who stumbled into legal by accident at a job fair, only to end up reshaping how legal teams think about tools, systems and scale. In this episode today, we get into what's really happening behind the scenes of legal operations. From navigating the chaos of AI adoption to cutting through the noise of all those legal tech vendors who say the same things. Aaron opens up about the changing roles of our ops leaders, of our general counsels, why so many conversations in our field feels vague or performative lately, and how we can bring something smarter and more strategic amid all this complexity. It's candid, it's sharp, it's a great conversation that I think you'll enjoy. So let's get into it from AI.

To M. And a change is accelerating. Stay ahead of the curve with exhibit A charts and graphics on the latest legal and market trends that are impacting your business.

Visit goodwinlaw.com exhibit A to check out.

Our data driven insights.

Hey Aaron, how's it going?

Hey, how's it going? It's good.

Yeah, good, good. I've been out of the in house side for a little while now. What have I missed? It's been pretty quiet on that side.

Oh my gosh, you've missed a lot, frankly. It changes every, it seems like every other week, but major, major shift from the normal legal ops strategy to now pivoting, really focus completely on okrs for the business. You've probably talked about it as well, but the GCs are expanding their roles, right? They're becoming more technologically savvy. I consider my GC a technologist. I don't think he would call himself that, but he is getting a lot of downstream ideas and requirements and okrs and so on from the CEO and elt. And so that's obviously being pushed down to the teams, especially the ops teams. And so we are still in the weeds with a lot of stuff, but it's much, much different now. So focused on the business as a whole. It's not that we weren't necessarily before, but it's just such a hyper focus on it now. And the what that looks like it would be, for example, how do we increase deal velocity? Right. So the component for legal obviously is, you know, negotiate deals faster in some respect. It's a qualitative thing. How do you measure it? How do you make it faster, better, easier? And so that's the first thing is I have to figure out how to measure that somehow. Right. And you know, you have clm, you have some other data points, you could start to put a story together. But it's missing quite a bit of stuff. Right. There's holes in that. And so that's actually one of the first things that changes. We really have to go plug those holes. Regardless of what it's going to take, you have to go plug those holes. And so very creative having to come up with new tools, data sets, a bunch of different things to try to figure out, look, how long is it taking us to actually negotiate a particular type of deal, particular type of paper, types of negotiations? It's a pretty massive effort. And so that's what I've been spending a lot of time on. And the kind of the internal legal stuff, how you help or assist the internal legal team, that's actually somewhat kind of taken a back seat based upon a team that I have, which is it's fixed. Right. And it's hard to get more resource. So we have to kind of pivot to that. And that's relatively new for me.

So that's totally true. So we have talked on this podcast and in other venues about the shifting role of legal operations and who your client is for essentially you and your team. And in the beginning, it is the legal team members. Right. You're trying to make their life easier, but as the role changes, it becomes much more strategic. You're starting to look at the rest of the organization. So the client of the legal team, which is the business as well as the leadership team, becomes more of the client focus for you all and thinking about how can you service the organization as a whole. Right.

I sometimes kind of feel like some ways, almost like an unguided missile, things are changing so fast that you know, you have a strategy, but then you realize your strategy's wrong or your strategy's changing significantly. What we're focusing on, what I'm focusing on my strategy is every single piece of data I collect, I need to expose it to the enterprise because GCs are communicating up to ELT. They again, as I mentioned, they want things quantified. And so every single piece of the stack, we are getting as much data as we possibly can. And to that end we created what's called a legal data model, which is a way for us to organize legal data such that it's not misinterpreted and misused inside the enterprise. Right. So someone might go search something in CLM and see like oh, there is an order form and they may say, okay, great, they're bound by terms of service or an MSA or whatever it might be when that's really not a complete picture. And so if we put all of this data in a way that's consumable and reliable for the rest of the organization, that allows folks to self service. Right. They can go build things off of that. It takes the burden off our back. Remember this is, we're, we're in an AI data cloud, right. And so data is incredibly important to us. And so we've done a lot of work to collect that. But one of the things we hadn't done is anticipated the voracious appetite for legal data. You'd be very surprised. And so we track legal users who did what work, who produced what asset, how long it took them and take. Like Sarah in Legal, she worked on a Salesforce case that included a CLM workflow for an NDA, an MSA and order form which generated 75 pieces of metadata. Right. And it took three months to close. Not perfect, right. But you have to start somewhere. All of that gets fed into a way that, like I said, is, is reliably consumed by folks that need it. Big, big shift, frankly.

And the metrics and the data that you're talking about is a lot more than just the old school reporting the weather, number of contracts reviewed, number of deals signed, et cetera. You're looking for things that can tell more of a story of a so what? Or as my GC used to say, how do we know we're getting better improvements over time? If you're doing more self service stuff, does the story come through? Are people able to see the so what? And that probably has to do with how you're surfacing the data.

Yeah, I would say that we are now training folks on how to use that data in the most appropriate way. And I think their minds are kind of blowing a little bit because what's happening now is the product, the service, the actual service, they are picking up our data and using its power features in the service. That's pretty wild, right? To be able to rely on that data and just within say 24 hours of someone signing a Document. It gets ingested into our tech stack and pushes out dispositions that folks can use to actually power Snowflake service, which was a big win for us. And frankly that was being used. Those are ideas that are coming from the other business units, not necessarily just us. And that's new. Very new.

Yeah. And how are you getting your direction? Aaron? You said your strategy, your team or your strategy and that's changed over time. That's changed from servicing the legal to servicing the business. Is someone telling you kind of what your okr should be for this year, what your goals should be or is that something that you have to scan the horizon for?

So there was a paradigm shift when we got our new CEO, right. Just a total different focus on how the business should operate. Right. And it was needed because as we morphed into this AI craziness it was Sridhar, who's our CEO really had that muscle. And so with that came his focus on okrs and stuff like that. So yes, we were involved, the senior folks in the legal team, they we were involved in creating the legal OKRs that matched some of the company ones. And that's where the company ones were very specific about how do we move faster and how do we release products faster and stuff like that. And so we keyed off of those now. Yes, they're changing. Yeah. You know we just finished Q1 and Q2 is slightly different and I think being nimble like that is. We knew that we were fast but being able to kind of pivot on some stuff as quickly as we're, it's been really refreshing and I think leadership is noticing as well.

And how does the team adapt to that? That's different. Moving fast, not knowing what you're going to be working on. And that is the environment that I've been most used to. I will say it's very different here than coming to a law firm that doesn't move in the same way like that software does. You've got to bring people along. But also I think it takes a certain personality type or like a work style to get comfortable with that. And it doesn't mean you can't do it. But there I often tell people the folks who will thrive in that kind of fast paced environment are the ones that are okay with the ambiguity and not having it clear you this is the priority for this next 12 month period because probably won't be after three months or five months.

Right? Yeah, I think everyone's trying to get used to this kind of shift. As I mentioned, it was Very different with the previous CEO. He was much more focused on out there in front selling, investor relations, driving investor value. And it's not the Trudeau is not doing that, but I think it's doing it in a different way. And so we're all still trying to get used to some of it, the pace and stuff like that, but I think we're adjusting. Like I said, this idea of an unguided missile. We really don't know what we're going to be doing in three months at least. I don't. From an ops standpoint, there's the core stuff, right? You have your bedrock, your CLMs and your CPQs and stuff like that. Those aren't going to change. But I'll tell you right now, I actually don't think I'm going to sign more than a one year deal on some vendor contracts.

I feel exactly the same way. I feel exactly the same way.

We get locked into some other folks and frankly they're boring. They just don't have anything interesting. We want to be the leading edge, right? And we're pretty risky. We'll definitely take on some stuff. And if we can't find something like that within it, because things are changing so quickly, right? If they haven't changed themselves within that timeframe, I think they're not for us, frankly.

Yeah, I totally agree. And I think providers are starting to get a little more used to that because it's changing too fast. I actually wanted to ask you what it's like to be leading legal ops in a top AI company in the midst of what's happening in AI. I know I can't keep track. Even if you see a demo a few weeks later, it's a whole different company and there's new entrants coming out. And I know your company is doing its own thing too. What's it like there? Are you looking at it constantly? Is that just a big theme of everything you do?

Yeah, it seems, I think every week almost we're now we're gaing some new feature product, whatever it might be. I'm not as dug into that piece, but it does have downstream effect. And so it is fascinating and somewhat overwhelming in some cases, just the sheer amount of stuff. But we have to do that, right? We have to keep up that pace in order to really perform in the market and to drive that shareholder value. And I think from my perspective on the technology that we're producing and stuff, it's become pretty easy for me to go build a chatbot, frankly, now to tune it and train it and all these different nice things you have with some of these larger models that you can buy off the shelf. That takes a while, but things are getting pretty good. The agents and stuff like that, they no longer focus on creating content, but more so about making decisions. And so that's really going to help because the decisions you had to make on your own to find the content that you needed, some of that stuff's going to happen for you. Right. And so we're looking forward to that flushing out a little bit more. It's nascent right now and I think it's going to take a little bit. But that's interesting. I wanted to get your opinion because right now, as I mentioned, I think folks are starting to build prompt models and stuff like that. Right. Starting to be able to offer their internal teams the ability to almost talk to some of these Gen AI things. Do you think that's going to change with the agents? Do you think that they're going to have to start pivoting to developing those roles versus prompts?

Ooh, it's a great question. I mean, I think agents are coming. There's no question. Right. And it's a very short amount of time where you have to start to actually think in a much more legal operations type way. Right. There's not just a prompt, there is workflow and process and step by step rules.

Yeah, I was just thinking about that actually.

Yeah. Does that change the way we work with it? It's a good question.

Yeah. Because you could use agents to build the workflows for you. Right. And so imagine you have a complex CLM workflow you need to build and you just have an agent go do it for you. That's the kind of stuff that I honestly don't know if that's going to be able to happen in two weeks, a month, three months. I know it's going to happen, but that's the tough part is how do I predict it? Do I want to be really leading edge? As a company that developed this stuff, it would be really cool. But I still have all this other stuff to do and so I kind of want to let some people try it first and then I can come in and obviously you have to create some sort of a strategy or use case for this.

Yes.

And that's what's different is we are, we have our classic use cases, right? How a CLM works, how to interface with other pieces of software like Salesforce or whatever. But then there's this kind of push for us to try These things before they're even vetted, right?

Yes.

Because a new feature comes out for us, it's readily available for us. And so there's this push to kind of say that go figure it out and go build something for us and even if it's not flushed out yet. And so again, that's very different. And I luckily have a technical background, but I can. If technical backgrounds right now are going to be so important for this industry. I mean, it's going to be. It's going to be a little scary at first, but I have a feeling that you're going to see a lot of former Comp Sci people or something like that maybe in this space.

Oh yeah. And I think it's a really attractive space to them because they see how much disruption and sort of greenfield opportunity there is here. These tools, I think you've called it in the past, like a new toy with no instructions. And that's, that's exactly, exactly what it is. I have said on some of my past episodes it's almost too overwhelming to show somebody like, here's a tool and now you have access. You should start using it in your day to day work. What should I use it for? And we go, oh, anything, just try it. And it's almost too much. And so as I think about, about how to roll something out, it's got to be probably a little more structured. How are you thinking about adoption and training and rolling out stuff like this?

Yeah, just in that point, as I mentioned you. I'm excited having conversations with folks these days. It's very hard to dig in tactically and talk about specifics and I. You hear some of the standard buzzwords and stuff and. Yeah, and so one of the reasons why I think that is, is because things are changing so quickly that it's hard to build that competency that quickly. Plus everything else is changing. As we just talked about the GC and even ELT is changing. And I feel like they're talking in these generalities because they don't really know how to talk about them in with specifics because it just. They don't know what the flavor of the frankly the week is right now. How we keep up with that is we, we have a talented legal team that knows they're negotiating these contracts and know these features. So they're a little bit better than most folks. But we are seeing folks that just go at Genai with no real kind of. They just say, does this agreement. What's the difference between this and how do I do that? And where are my contracts that do this? And they're, I wouldn't say failing, but they're struggling a little bit. The prompts aren't really flushed out. What's good news is I. I've been seeing playbooks out on the different communities that are exchanging those prompt libraries and I think that's gonna be huge. Right? Because I might use them too. I might just take them and put them on one of our internal knowledge bases and say, have at it, go have fun, use and see what you can find. Because again, self service is important as we're a pretty big legal team now and it's hard to. It frankly is hard to fulfill all the requests, stuff like this. And so the more we can self service folks with that data I was mentioning with these tools, on top of that, I say have at it. Right? It's as the kind of the leading edge of this stuff. It's still hard for us too. Resources are constrained because imagine everyone in the company trying to do all of this stuff, just like we're doing, to see how you can make something that's better, faster, easier for the team. So a challenging time from a resource perspective. Incredibly exciting though some of the stuff that goes around in our slack channels and stuff like that is amazing. So.

Yeah, it is.

We're hanging on, if that's to answer your question. We're hanging on.

And like for the prompt engineering or the prompt writing, how big a gap do you feel like it is? Because there's vendors that I talked to this week that provide training on this. They teach people how to write a proper prompt. I think back, I'm not sure it's a fair comparison. In the old days of trying to learn how to Google for something and there'd be people who you're just like, get outta the way. Let me. You didn't write that, right? Here's how you're gonna get to the right blue. First blue link is the one that you want. Am I feeling lucky? Cause you know exactly what to put in that bar. And that took, I don't know. No, no time for people to figure out pretty quickly. And in my own experience, I remember when ChatGPT came out and everyone was like, what is this? And you can like write all these silly things and it can write raps for you and you didn't know what to do with it practically. And if you. That muscle of knowing how to start asking it to do stuff that is more sophisticated. If I needed to use it for work, like I can be pretty successful I can write a pretty long prompt and get it to do what I want it to do. Do you think that's something people will just learn very quickly? Like we all learned how to do a Google search? Or. Or is there actually something you need to be formally trained on here?

I think you're going to have to be formally trained on it because if you think about it, you have to tell this agent whatever you want to call it about yourself, about what you're looking for. I'm an attorney. Here's a specific use case and it needs to be measured this way. And it has to do X, Y or Z. That's really hard to do without really knowing how to build that. So yes, I think they'll learn it, but I definitely think that it will take, frankly it might even take some external consulting or something like that because if I don't have the time, perhaps they can help as well. And like I said, I couple hours I could spin up a chatbot in the snowflake environment because it's that easy. And you could bring your own models. You truly is amazing. But I think the folks won't know how to use it when they do because what I see is the ones that have started to try to use it. It's like a sentence here. It's a. You're. They have a conversation. Right?

Yeah.

And so instead of saying here is the universe of everything up front, we can talk later. Right. But for now, this is what everything's going on and I want you to really understand the background and have a mission almost. Yeah, yeah. And then we can chat after that and we can fine tune and filter out. And I think that's what I'm seeing is just like one liners here and there, just trying to dig through. It's not giving the whole picture. And this is the challenge with legal. Right. You can't just, you know, fork over 80%. Right. You kind of have to be pretty close to accurate because if you have a security incident, you're trying to find all MSAs or security denims that have XYZ in them. You need to be pretty close to finding all those. Obviously. Sure, there's going to be some mist here and there. Most likely for us that percentage needs to be pretty high for us to really trust it. And I think that's what they're seeing. They'll write a prompt, they'll one or two lines and then it'll come back with very much less than what they're looking for. Ah, you know what? This thing doesn't work. I've seen that so many times.

Yeah, I find that fascinating because I would think that what you learn is, oh, I didn't give it enough context. And so let me write you like a whole page worth of what I would explain to you what I'm asking you to look for this back and forth because I agree the back and forth is not successful.

Yeah. I almost think that there's going to be an agent to go talk to you about building your prompt or something like that with the purpose of building something that's useful and contextual and all that. That's very possible Again within three months. It might read your mind. I have no idea. Frankly, it's. It is pretty fast. I was just at the Snowflake Summit Conference, right?

Yeah.

And that's where they show the new cars and the shiny new toys and all that. Even more fascinating than I ever thought. It could be just a paradigm shift. You could already tell, like just things have changed so quickly. For example, the big kind of main keynote area. Somebody said, please generate me or compose me some music or something like that. They got an orchestra and then they basically played the composition right there in front of everyone and it was beautiful. It was fantastic. And then they wrote, okay, increase the sound or increase the trumpets or add more, I think softness to it. I forget what it was. And you could actually hear them starting to change the actual. The orchestra was actually changed. So that was pretty amazing. And the fact that it could do that so quickly and the fact that you could do it on really any model you want, which is, I think going to be. That's going to be interesting actually. But your data in these legal tools. Right. If I can find a better model to get at it and use it, I think folks are going to have to let them do that in a lot of ways or offer them something to let them do that. You hear folks that are starting to add on multiple models, Gemini, OpenAI, whatever, then they're building something on top of that. I maybe there's going to be a point where you're going to be able to build something on top of that on your own. Because a lot of folks aren't necessarily. They don't have the tools that we have, which is to take that data and do stuff with it. They may have to bring something to that app directly. And I think if that app doesn't give them the ability to do more, if I was them, I would. Because then you're almost self servicing on some of that stuff as well. Maybe they've developed some in house way that has more context and apply it to some CLM model that's AI power contractor for you or whatever it might be.

Well, each model is in my experience, better at different things and so you have to think about what the job is to be done. As Jake Saper would say from my last episode, let's talk about the vendor maze. I mean yes, you can spin something up internally really quickly. You can buy a bunch of things. The things that I see that look interesting by the time I've started to think oh, maybe we should pilot it. Another providers, we've built that into what we're doing already. It's what on earth, like how do you even decide which one to buy when there's like point solutions, there's enterprise solutions, there's legal solutions, there's stuff that's coming out of the existing SaaS that you have. There's stuff coming out of the providers that don't do anything near that, but they can do it too. It's like, it's overwhelming.

Yeah. The clock landscape was interesting because there was a bunch of Word plugins I think I was telling you, I counted 20 or so. @ least some of the bellwether ones have it of course and they're fine, they work great, but they work great for probably most folks. And I think those are, to your point, they're bringing that kind of stuff into almost every single concept you can think of. For example, we have an app. It really does red lines really well from, not from a negotiation standpoint but comparing a bunch of different file types and stuff like that. So they just added something now where you can plug it into Word and do all kinds of gen AI type stuff, ask questions of it, do outside research. Pretty fantastic, right? We never would have thought that they would have added that ever. And so I think it's a. I don't mean to trivialize it but I do think it's low hanging fruit. Right. I think it's become so easy to build this stuff that and so many tools to be able to do it. I think it just seems natural for the folks that are established in the market now, the new folks, I sat with a few of them and Mary, I don't think they've ever sat with legal teams for an extended period of.

Time. Oh yeah, I.

Agree. I feel like it's. I think I told one of them and it's a very lazy approach. Right. Because Snowflake has a. We have a playbook. Right. But it's issue based. It's not just pieces of language and stuff like that. It's, look, here's a customer situation. How do I handle it? How do I negotiate it, the context of it all? And you know what, there might be a sentence or two here that we want to add in. Right. How do you do that in a word plugin? So what? To answer your original question, we are pocing a bunch of stuff basically. And so I think it will be one of them might win. None of them might win. Don't know.

Yet. It's so hard. Every single thing that I see I find useful. Yeah, we could use. That's useful. There's a use for that. There's impact there. People have a place where they're used to doing their work. Maybe that's why we see so many word plugins, right. We don't want them to have to go to another, yet another tool and yet another portal and another window in order to get that job done. Do you end up buying like a bunch of things that do little things really well? Does it end up being one of these incumbents that just are where you work? Does it all become like a chatbot? No, I don't know.

It's. I think the first thing you could. You should do is go to your. Go to your trusted vendor, whether it's your CLM vendor or something like that, and say, where are you at? How. Where are you going to be? How do we get you there faster? Be part of that product team if you need to. Really offer as much as you can from your skill set. And if that doesn't take traction and you don't see that on the roadmap. Yeah, I think you could supplement it because I think that the attorneys at minimum will find the ability to clean up the documents or have their playbook, their language, or at least east, at least their precision red lines and all that kind of stuff is pretty straightforward. Now, a lot of them were touting that, but they all did it and they all said, we're the only one that we that does it. And of course, most of the other folks on the floor did it as well. I don't necessarily know if the negotiation piece is going to be as useful as right now, at least right now as like, look, here's all my stuff. It's more efficient for me to just move stuff in if I want. Here's this. Oh, and here's the issues, right, that you're spotting. Just forcing it into that and just allowing folks to access it. But this idea that kind of stitch it all together. Because at Snowflake we are experts at first party paper. Right. We're going to write the language. We've written I don't know how many times. Right. And so third party paper is much. It's definitely more rare. So I think a lot of this is actually going to really help the vendor landscape the buy side. We get a lot of bespoke documents. So if we can have something that helps us understand all of these different positions that came from some agreement, like a hotel agreement or. What's funny is when we released Snowpark, which was something to do with Python, the language, we actually got a Python, a real Python and they wanted to wrap it around. I forget who it was. The agreement for that legally was apparently incredibly complex liability, all that kind of stuff. Right. What if it bites in the face? And so that kind of stuff. How do you, what do you do? And so some of these apps, which I think is really cool, is you can actually go do outside research. So for example, let's say the customer includes some odd regulatory reference that you don't know about. The ability to buy, as you mentioned, to do things in the same doc or sorry, in the same app to be able to just go research that regulation right then and there. And then you could go ahead and respond to it with some sort of negotiation tactic. But I think that there will be. The buy side will definitely, I think, get more out of it than the folks that are doing the same thing over and over on the sell.

Side. And the vendors really need to be sitting down with legal teams, as you said earlier. And they have so much to learn from what actually happens on the ground and what the jobs.

Are. Did you see more procurement teams in legal or outside of legal contract managers? Did you see those in legal? Not in legal. And in your experience, procurement legal for.

Contracts. But the procurement folks usually did not sit within legal.

No. Yeah. So what I mean is like we have procurement contract managers, but they do, they do not sit in legal. So they are attorneys, but they don't sit in legal. Then we have a procurement legal team which are escalations from.

There.

Yes. And so that's common. Yeah. And so it's. It's a little weird in that we have to go to each to try to figure out what's going on with the deal. But I do think that the contract managers over here, that first line of kind of triage, I think that's where these tools are going to really going to help them because it helps to get through these. I think Procurement folks are just bombarded with so much stuff. If they can build a playbook based upon the stuff they've already done and used, that would be.

Fantastic.

Yeah. And I think that's available. It's just really hard to find this stuff with so many players out there. How do you wade through them.

All? Yeah, I think you do a.

Lot of POCs, but that's so much work.

Too. And I want to go back to what you said earlier, which is when you talk to all these folks, the definition of what they do is very shallow or.

Vague.

Yeah. Is that true? Was that always true? Is that a new thing? What's missing? How do we go back if it wasn't the.

Case? I get the sense that we used to be focused on the legal team. Right. How do we make that legal team run as optimized as possible and build them the tools and stuff? Get those tools. You could try them out, you could test them, you could create your RFPs. You can do it. Now, that being said, larger teams can do that. They still are doing that. I totally get it. But the nimble teams that we have finite resources and are moving incredibly quickly, we can't do that anymore. And I think the people like us who are experiencing that as well and who don't have that technology or technical acumen, that those are the folks that are just, okay, we need to optimize something or become more efficient with something. And this is where I think you're going to need to see more consultants coming in and helping the legal ops teams get that, that muscle, as we talked about, because I don't know if they're going to have enough time to build that on their.

Own. So you're saying that description of what we do is not. You're not saying it's the software companies that are unable to articulate what they do, but it's actually the legal ops function and the people in the roles who can't articulate it.

Anymore. That's what I'm seeing. Yeah. I think it depends upon, again, which. But the larger teams and stuff, they still have their classic. I think they call them centers of excellence and.

Stuff.

Yep. We gotta be excellent for everything on our stuff. There is no center. It's just. It is. We just go. And someone picks up the detritus that we're leaving on the.

Packet. Yeah. The role. Is it vague? Because it's changing. We've talked about kind of some of my observations in the field, which is we saw a moment where there was an explosion of roles that Were very strategic, very high level, large companies, really broad scope like and growing scope. But I feel like there's been a fall off on that. I don't know that everyone is really clear. There was a point where everyone started to get much more clear on what the role was and every GC wanted it. And I think it's going backwards a little bit. Maybe it's because the world is changing so fast and the needs of every department are just unclear. And so it's just you need someone to do stuff, you don't know what it.

Is. Yeah, I think that's exactly right because the GC is going in and walking into a CEO. It's like hey, what's this Gen AI stuff? Go figure it out. How is legal applying this new technology? Okay, so first thing I'm going to do is I'm going to put a rec out for some Legal Ops person. Right. Personally I'm seeing quite a few Legal Ops positions coming on the market now. Again we have this pay issue right where some of them, they want 10 years of experience and $100,000 in salary. And so you can tell those are the folks in my opinion that don't understand what they're getting into or what they need to get into. In a lot of ways I have a feeling if I was a fly on the wall, those are the ones where it's like go figure it out, go find me some Gen AI stuff. And they're just flailing around. And I think those that dollar amount's going to attract obviously folks that are probably more early on in their career which I think is just going to perpetuate the problem because that's moving so quickly that they're just going to get eaten up. And so I think the folks that have a mature job description with an adequate pay and you can really tell that they've thought it through. They still need the outside counsel stuff, they still need the KV stuff, they need that. But I've seen on almost every single one data, data, data, which is perfect because it's exactly the mindset you're need in order to compete and to be, when I say compete, be a really world class legal team.

Yeah. And data and AI seems to be a big focus now of everyone. Does that change the profile, the background, the skill set that is needed to be successful in these.

Roles? I think so. I've seen a few new roles that actually want folks that have some sort of a technical background. And I think the concept, honestly I think if you're a really good ops person and you have some technical experience. I think some of them might teach you the legal part.

Right?

Yeah. And that's a different. And at least for me that's a little bit different than. Because you usually see paralegals or whoever kind of move or mature into that role. But I have a feeling that it could be consultants, could be someone MBAs. I think almost every one of them, they want an MBA. I'm seeing less JDs. Are you like right? I am seeing. Because usually you see like nice to have or they can be a jd, MBA or whatever. I'm starting to. I don't see as many JDs, which I think is.

Good. That's refreshing. I know you and I are aligned on just that. The skill set needed. You can learn legal, but you can't learn the business skills and the project management and the data and the analytics and all that. And maybe that makes it, as we're saying, the profile of the role changes such that the paralegals and the lawyers that are moving into this aren't necessarily the most like most well aligned to shift into this role versus someone who's come from a totally different field and then just has to learn.

Legal. Yeah, we actually hopefully are getting ready to hire someone who has. He works at an ediscovery company, but data engineering can program business intelligence stuff. Dashboarding vision. That's it. That's the function that would have been unheard of a year ago. Right. What do you guys. You're creating your shadow IT team over here. Right. The problem is there's not enough resource and so it's so important upstream to folks that you can do it. You can actually, if you work hard and you make the case for it like we did, you can bring in your own kind of data ops person. And so that is going to just be a godsend because mentioned we're going so fast it's hard to articulate strategies and document them. And even BRDs are becoming a challenge to try to get out. That person can take all that off of your plate and you could free you up to maybe go back as I mentioned internally and fix some problems or help folks just onboarding folks these days, like you're just powering through it. Like are you good? Figure it.

Out. I know it's so fast. I've just brought in someone who's going to be a very important role here and who came from outside of the legal space and I gave him a couple weeks for him to just kind of learn things and then we sat down and I Said, okay, this is how a law firm works. These are all the words, these are the structure. This is the financials. This is how we generate business. This is the revenue model. And it was like about an hour, an hour and a half of education and it's like okay, I got it. It's not that.

Complicated. Yeah. And that's what if you hire the right person that operation mindset. That's probably a good way to put it. I'm seeing the folks that maybe we're losing that operations mindset. We're getting to a I know how to do legal stuff. You know what I mean? But I don't when that really the rubber hits the road, I'm not an operation necessarily. Those are the, the pieces that I'm missing and maybe that's what it is and.

It'S. It's hard especially for the first hire. So have so many of these conversations. Spoke to a GC just this week of a hyper growth startup. We have an enormous volume of contracts and we just can't get through it. And I need a legal ops person to come in here and get my CLM going and start putting some process in and making that turn into a machine. And that's the number one job of this person. I was like okay, that's a really different skill set than if we're talking about this is the person who's going to help you think about how to right. Source and strategically grow and position and the firm and or the company, et cetera. And I go do you want someone to be able to do this tactical stuff and grow into that? Are you then going to bring in someone to layer over them? Like how are you thinking about the legal ops as a function and how does it evolve? And it was like and I understand I can't see past the next six months. This is what I need right now. And if this person can't figure it out then I'll deal with that.

Later. Yep. Yeah, I totally.

Agree. You see a lot of that and the advice of then how do I grow the team after that is another gc. I talked. You said okay, we have a steady state now with our baseline tools. How do I move to data analytics and a much more strategic function with the people that I.

Have? That's.

Yeah. Might have to bring someone else.

In. The space for visualizing data is exploding. That is tough for me because I don't know which one to use. I know some folks just want pretty pictures. Other folks really want deep analysis. Does one of them do both? Is one of them too complicated for tableau does great, but it's a really steep learning curve. And then you have other ones that just may make pretty pictures but they don't do enough. This fine balance in like CLMs and stuff like that and other apps is the more functionality you add, the lower the usability, in my opinion. So you need something that's right in the middle there where it does enough so the folks like me and the folks like the operation mindset folks can use it and they don't have to go to IT for everything. Right. Because we are also strapped for resources that we almost need to take control of our own. The ops teams at least we need to build things that allow us to service those things as well. Now I think smaller companies, more nimble companies, they can, they probably have that, that situation or scenario. We're 8,000 plus people now, right. We have an IT team that's gigantic. So we can't all just go be buying things and doing our own stuff. Right. We need to be in sync. But like I said, there's that lane where audit lets us do enough. We don't break any kind of revenue controls or something like that. And those are almost like unicorn apps, right? And we found a couple of them and so I think there might be more. Hopefully there's more of those especially in the data visualization space and these.

Kind of data operations and what you're explaining, everyone moving in sync, moving in the same direction. So much speed needed, so much coordination needed, but also want to get out of each other's way and not be a barrier to getting stuff done. That is a very important just like leadership mindset, cultural like dance between all these teams. We had our partner retreat recently, had a really great speaker talk about that where in the old way of doing things, everyone has their clear role and responsibility. You put your head down, you do your job, you stay out of everyone's way and you do like your job description. That doesn't work anymore because we're moving too fast. There's too much work to be done and we can't be territorial about it. We can't wait on each other. And so you have to almost have a clear vision and direction for the entire organization to move in. And everyone knows we're all in one team. It all needs to get done. Holler, right? Tell me, tell me where you're going. I'm over here. I got this, you got that. And it all just has to move. I'm also observing the traditional way you mentioned. Do we have Time for a brd. And then here's like the six month project plan and here's the ROI analysis and here's how we're going to roll it out. No one has time for that anymore. And it almost feels like overbearing and too structured. Yeah, yeah. And it's overbearing. Right. Don't give me like a 12 page PowerPoint and Gantt chart about like why we should do this. Is it a good IDE idea? Let's just, let's go. And how do we get there without tripping over ourselves and getting in trouble but at the same time.

Moving. But you're gonna break stuff. Right? But it's okay. I think one of the things that we hear a lot internally is fail fast. Right. And so that's what we do. We just have to. Because we cannot sit around idle and wait for stuff especially, especially now like the time we're in. So I think that we will see where this fleshes out. But I have a sense that if we can get folks either trained up or operations minded folks like you said, leadership roles who know how to play nicely with other teams. Because I know it's very hard at least for me to say hey, go ahead and here's what you need to do. Go ahead and do it yourself. And it's hard for me not to just say my team's going to do it because that's an ops type thing. I've had to get over that. Right. I have had to say I can. I'm going to teach you how to fish and you guys go do it. Right. Because I would. We have teams that are going around the classic legal ops function to get their stuff done at least in legal. And so we used to say stop, there's a queue and you got to. But now we're kind of like look, just nobody kind of kill each other over resources. But let's all try to stand up our own little projects and go for there. I. It's. That was very tough for me to overcome. Yeah, you're losing a little bit of control over. You're very well defined. I have a process and I can't really have a process.

Anymore. In some ways that's right. I think that's hard for everyone. And if you're doing something more big cross functional like rolling out, I don't know some of the dashboards or the reporting or the new tools that are self service as you mentioned. How are you going about doing that? You said you don't have a ton of resources but you can't just blast it out there and hope for the best. Are you thinking through without a lot of resources how to do that? Because it doesn't all need a ton of training anymore. But it's.

Adoption. The key thing for us though is that we want to. There's Snowflake on Snowflake kind of mentality. And so we do offer some BI tools in the Snowflake service, but that requires folks to help us. It's I can go in and write some cool SQL queries and stuff like that, but now you're talking about almost like building some apps and stuff and that I need help with. And that is where the resource comes in. And so we actually don't really go by outside. At least legal. We don't use outside visualization tools. We try to use whatever the company's using. You get to an enterprise level company or sorry, enterprise level apps that you're purchasing third party apps and it becomes you are required or at least you're, I would say obligated but you should be a good corporate citizen and use those tools. But they're just, they're not a good fit like I said in that. That one kind of sweet spot where we can do stuff on our own. So I'm making the case obviously for the budget and stuff to be able to do or to buy those type of things. But again it's kind of look we are going to have. If everybody gets their own visualization tool, how are we going to maintain it all? Licenses, money. I understand the challenge, but we have to do something right. We have to continue to grow and build the team in a way that makes sense for what ELT wants, what the GC wants, that vision. And if it means that we have to spend a little money to get what we want, I think that's going to be okay in the coming future, very.

Soon. Yeah, that's really hard. The enterprise tools and what other teams are using and how do you coordinate? That's why we don't buy for more than a year. Right Aaron? Yeah, you fix it.

Later. Yeah, of course. Don't go buy another CPQ or something like that. But like outside counsel stuff, one year deal. I want to see what people are doing. There's got to be something cool that's coming out. Even the kind of ask questions about your reporting and stuff like that's already boring in my opinion. There's. I want to see how do I get stuff out of the attorney's hands better, faster, easier, how do I help with some setting up panels, all that kind of stuff like. And I know there are companies out there that are doing that, but they're, they seem to be not all in the same place, which is unfortunate. I don't really want to go get three or four things to, to do my outside council billing. I. I would rather see what I can get from folks and then see what their product roadmap's incredibly important. That's the first thing we look at, is where are you guys going? And if we don't see anything interesting, we usually move.

On. Yeah. Aaron, where do you think things are going from.

Here? I don't know. I think it's going to be a very tumultuous next six to 12 months. I really think so. I think that we're going to have turnover for the folks that are struggling, because I have a feeling that the GCs are going to get way smarter about this. They're going to revolutionize their own teams. Right. And so I think if their ops teams don't keep up, I'll be honest, like, I feel sometimes I'm falling behind in some ways. Right. And so I'm really focusing on, look, Dirk is our gc, like, what can I do to help get what you need in order to make this company go as fast as it needs to go? And it may seem very strange that these, some of these obscure legal functions have anything, any impact on that, you'd be shocked because it all starts to add up right, at the tactical level and then you actually start affecting change upstream, which is really fascinating to see and it's quite rewarding. But I don't know if a lot of folks are going to get to that point. The ops folks, right. I feel like they're going to have to move very quickly and learn very quickly because their teams are going to probably outpace them because they have this voracious appetite to learn, because they have to report back up. Right. And so even my GC is like, hey, I saw that. I saw they forwarding stuff to me, like, what do you think? What do you think? And you know what that means? That means go check it out, right? So you're constantly looking for that new thing, that new shiny feature, and that used to not be the case. It used to be where you had a very methodical, proactive approach to buying.

Something. I.

Agree. It doesn't seem that way now. It just seems, look, let's just pass the net out there and let's see what happens. So where we're going, if anyone has a problem with that if anyone has any, any kind of pushback on that, I think those folks are going to struggle. I would love to see the current leadership be able to adopt this new mindset of data and how do I apply it and how do I try to quantify some of the legal stuff? Because I understand majority of it's qualitative, but if you can start putting together a story, think about it. If you have that data and you say we need a new commercial attorney in Singapore because MSAs have exploded, the sales rep to legal ratio or attorney ratio has skyrocketed. All that data came from your stack. Right. And so that's going to become super powerful. And I think that's what GCs are going to.

Require. But I think you made a really interesting observation, which is the GC is getting smarter. They're getting more strategic, they're getting more sophisticated when it comes to use of data. They're getting much more aware of what technology can do to modernize the way that work is done and legal services are delivered from their team. And the role of the GC and their changing role has always had a really fast and direct impact on legal.

Operations.

Sure. They do what their role is, where they sit in the organization. And so that is all changing really fast because it used to be that the GC had no interest in a lot of this stuff and kind of threw it over to another smart white hand. But now they're getting smarter and so that starts to change what they need from their right hand, lots to keep an eye.

On. I am shocked at how quickly RGC has. He works at a very complicated, sophisticated AI, so he understands that stuff. But the ability to recognize that something might work for his team in legal has gotten so much more acute. I really. It's completely noticeable. I. It's could be frustrating and stressful sometimes. I wouldn't have it any other way because I love being in the middle of this revolution. And I think if you're not interested in that, if you just like the way things are, I don't know, it's going to be.

Tough. Yeah, I think you nailed it. Every GC I've spoken with in the last six months seems to have a much stronger grasp on the vendors, the players, what's happening and asking really good questions. And they maybe weren't being pushed to do that in the.

Past. No, they.

Weren'T. We'll.

See.

Yeah. All right, Aaron, we'll keep an eye on this. But such a pleasure. Thank you so much for being here. It was great talking to.

You. It was great talking to you, too.

Cheers. All right, everyone, thanks for.