[00:00:00] Bill, thanks for joining me this afternoon Hey, Ted. It's great to be here. Yeah, man, we've talked about doing this for quite some time, and I'm glad we f- we're finally getting it done It's, uh, it's g- it-- You know, I love cha- I love talking with you, and now we're gonna do it, uh, in a way that lots of other people get to listen in too. Exactly. I mean, that's how the podcast started was I-- when we, we used to be a company called Acrowire. We were, like, a systems integrator for years, like 14 years, and then we productized, and I was re-introducing us to the market, and I was recording my calls, and I was going back and listening, trying to refine my messaging. And I was like, "Man, these are really good. Like, this should be a podcast." And, uh, 130 episodes later, here we are. Yeah. You've done, you've done an awesome job with it. I appreciate that. Well, why don't we get you introduced? Um, why don't you tell everybody, you're, you're [00:01:00] like a legal tech OG and have been around for, for a long time. Why don't you tell everybody a little bit about your background and kind of what you're up to today? Yeah. So like most, uh, 18-year-olds, I decided to write an accounting system for law firms, and that turned into a pretty good, uh, start of, of a career. That turned into a system called, uh, ProLaw Software. We were successful enough that, uh, we were able to sell the company to the West division of, of what's now, uh, Thomson Reuters. And I got to do all kinds of wonderful and amazing things, uh, because of that, including, uh, creating the, the West KM, uh, division at West, built another, uh, KM, uh, system called, uh, Exemplify, which sold to Bloomberg. So both of those are, are AI products. So I've been working in AI in legal since the late '90s.[00:02:00] So I, I guess I have admitted how long I've been doing this. And, uh, and then came back in, uh, after selling Exemplify in, I think it was 2016, came back into legal again in 2019 to, uh, to start Zebra Workz, which is, which is how we get back, uh, back to here. And, and, uh, what does, what does Zebra Workz do? So when, uh, uh, when my partner Ken Bassam and I got back together, we said we should, we should come back into legal, and didn't know exactly what we were gonna do. So we went out and just started talking to a lot of firms that we had been working with. We wanted to get back into the finance side, which is really where most of my experience in legal has been. Been working in finance for, you know, forever in this area. And it was just this sort of consistent thing that kept coming up over and over again, uh, because there's been a lot of investment that's occurred in the core part of let's track time and let's get the bills out. But then there's this rather crucial [00:03:00] stage of getting paid and how do we increase our realization. And, and the themes around that just kept coming up over and over again. And so we just saw a big opportunity there. And that turned into creating the first legal specific invoices to cash system, and that's, that's what Zebraworks is. Interesting. Yeah, and I remember ProLaw. I mean, it, it's still around to a certain extent. Um, and y- there was even a DMS as part of ProLaw, was there not? Well, ProLaw was very unique because we really created the first integrated practice management system that did both the back office and front office in one system. So it was document management, docketing, case management, along with the full accounting, billing, and so on. Interesting. Well, um, you and I have talked about AI and how this industry is managing its way through this transformation, [00:04:00] and I think most recently at TLTF, uh, we were chewing the fat, if you will, and talking a little bit about how all of the focus, uh, not all of the focus, maybe I'm exaggerating a bit, but probably not by much. Much of the focus with AI implementation is really focus, focused within the practice. And I get why it's, it's where, you know, that's where the opportunity cost is the highest, right? If you can save a lawyer an hour of time that he can then turn around and bill to a client, you know, that's hundreds of dollars, maybe even thousands of dollars for that one hour of, of time. So that makes sense, where in business of law scenarios, um, that opportunity cost isn't quite the same However, there are many compelling [00:05:00] reasons why I think, and I believe you agree, that starting in the business of law functions makes a lot of sense. And we're kinda-- It, it's a little late to talk about starting. If you haven't started yet, um, you're in trouble. Uh, but, you know, I think that y- I've been beating this drum for a couple of years now. The, the cost or the risk reward is even though the reward side isn't as high as within the practice, the risk side of that ratio is so much lower. If you make, if you send out the wrong, the, uh, invoice or, you know, um, y- the applicant gets a, you know, in a HR process, like the stakes are lower. It, it-- Anything external facing is not good when you make errors, but it's like exponentially worse [00:06:00] within the practice, right? I mean, we, we-- we're seeing this play out in real time. You're seeing very reputable law firms that are delivering, um, documents to the court with hallucinated citations and factual errors, and it's just really embarrassing, I would imagine. So, um, I've been beating this drum for a while, but there seems to be a disproportionate focus on practice of law solutions. I guess let's start with, do you agree that the business of law is a good place to start? It's a wonderful place to start, and I would push back a little bit on the reward part in the sense that it's such a direct connection between putting AI to work in the business of law and the ROI that you get from that, where in the practice of law, you're talking about fundamental changes in just how the firm operates, the service delivery. [00:07:00] Whereas in the business of law, you're just making the process more efficient, and particularly in the, in the area that we work, the results that you get from that have a just absolute direct ROI benefit. So if you currently have, you know, if you're a large mid-sized firm and you have total, uh, lockup of 150 days, and you bring that down to 110 days, that 40-day reduction in AR lockup, think about if you are doing 200 million a year in, uh, in billing, that 40 days, what it does in terms of the increase in working capital, that's, you know, that's a real reward. And a natural result from that is you're gonna increase realization, and so you could easily get 150, 200 basis point increase in realization by reducing your lockup by that. And that's, that's cash that was, uh, just going to disappear. So it's, it's straightforward. The [00:08:00] results are very real, and it's just business process that you're making more efficient. You're not trying to fundamentally re-engineer what a law firm is, which in theory is, is what AI in the practice of law has the long-term potential to be. And, and we can have all kinds of arguments, and you've had many podcasts on exactly that subject, which is why I think it's so fascinating to talk about this other, other side. Uh, my personal opinion is it could take a whole lot longer for the practice of law side to have all these changes that everybody is so excited about. And in the meantime, this is an amazing opportunity to go focus on making the business of law more efficient using these same core technologies for a very direct and beneficial result. Yeah. So we, we agree. We're, we're aligned on that, [00:09:00] and yeah, you bring up a good point. The The bottom line impact, unless you kind of re-architect your pricing model and a number of other things, your internal firm compensation model, like lawyers today are, are judged on top line performance. Very rarely are law firms rewarding lawyers based on bottom line performance, right? Which if you're in a scenario where you're increasing productivity, the bottom line impact is where you're gonna see it if, if you're pricing appropriately, right? It doesn't really work out when you're billing hours, uh, unless you're just raising your hourly rate, which there's a s- there's a ceiling to how far you can go, even though every year it seems like we bust through that ceiling. Um, it's, it's actually pretty amazing. But, um, you know, [00:10:00] the-- Here's what I see. So Info Dash, we are a, um, w- we're really a legal innovation platform. We, we sell intranets and extranets, and we have a AI enterprise search, um, platform. Um, but we are the delivery rails that you can use, that a law firm can use to deliver solutions either internally to clients or, uh, internally to stake-stakeholders, employees, or externally to clients. And on the extranet side, we compete primarily with HighQ. And HighQ was bought by Thomson Reuters in 2019. Very little has changed about that product except the price in that time. And there's a lot of, th- there's a lot of appetite for alternatives. Um, however, what we have found is that even though people don't love it, and I'm, I'm being kind here, I don't wanna, uh, I don't wanna bash other vendors, um, but [00:11:00] there is limited, um, there is less motivation to rip it out and replace it because right now, like our primary ICP is we sell to chief innovation officers, chief knowledge officers, and they're getting a lot of pressure from leadership to get AI wins on the board, right? So am I gonna take a system that I don't-- is overpriced and I don't love, um, but it works technically, and replace it with something much better and even if I save-- Or am I gonna go focus my, you know, it's like you're playing craps and you got certain number of chips, and you gotta figure out where on the table you, you distribute those chips, and you only have so many, and that's your change management capacity. So how you allocate that a-align, you have to align that with how you're gonna be measured. And, um, I-- it seems like that is a significant constraint in the conversations we're having. Is it, [00:12:00] are you seeing it the same way? It, it is, and I think another factor that really plays into this is that when you look at the staffing in firms from pandemic forward, in essentially every area of the firm except for finance and accounting, firms have gotten more efficient. The one area where there's just a consistent increase in FTEs is, is in exactly what we're talking about here, and that's because this area just gets increasingly more complicated. So the challenges with OCG and the constant issues with, with, uh, keeping up with billing requirements and pricing and all of the challenges that come around running the back office of a law firm, the way firms have been dealing with that is, is to throw, is to throw more heads at it. And to an extent, that works, and from an ROI perspective, it works also in, in that putting the right effort into [00:13:00] getting the pricing right and the right effort into collections. Like, every time I hire another collector, guess what? We get, we get more money that comes in. That makes a lot of sense. But there's a point where the process and the technology and the platform that we're running on, more people doesn't, doesn't fix the problem anymore because the process is just, is just out of whack. And, and so I think because that's taken such an investment and it is so apparent now relative to all the efficiencies that have been gained in so many other areas in the firm that it's really kind of put a spotlight on, on this issue. So that side's been very beneficial for what we do specifically in, in finance. The thing that you're talking about, I, I think really highlights an important point, which is that just cost savings is really not enough for a firm to go through the [00:14:00] effort in switching technology because the real cost is, is the time that it takes to make the switch, the retraining, all of the, you know, all the ripple effects that occur with making that change. So it's not about we're just gonna get a better solution and it's gonna be lower cost and things are gonna be better. We need a fundamental order of magnitude change that's gonna occur because we are doing this. And in your case, that's because we're getting a platform that fundamentally changes our ability to then innovate moving forward. It's not about replacing HighQ, it's about where are we gonna be after we do that. In our case, it's about being able to fundamentally change what the velocity is of the billing collections life cycle in a firm without just throwing more people at it. Right. You know, and a, a gap, uh, that has existed in a, for a very long time in the legal [00:15:00] industry is really around process engineering. You know, we, we talk a lot about technology. I- to scale a process with technology, you need to have a certain amount of standardization. Um, l- law, especially within the practice, has been very bespoke. And, um, you know, I, I'm a Lean Six Sigma Black Belt, and I-- One of my first legal clients, uh, a law firm called Hunoval Law in late 2000s, they were a default services firm, um, started out with just a handful of people, absolutely exploded. As you can imagine, there were a lot of uh, foreclosures happening in that timeframe. And we-- They were almost like one of the first AI native law firms without being AI native. Um, they were kind of process native. So we, um, I ha- me and another, [00:16:00] uh, another black belt who I worked with at the bank, I pulled him out, he came to work with me. We got embedded with this firm and scaled them from, I don't know, 25 to 500 employees in, like, two years. Like, m- pretty massive growth. Still a small law fir- in, you know, law firm, um, scales. But-- And we built-- Uh, they had a BPO, so they had a partner in India that did a, a lot of the grunt work, but not all. And we s- we organized them in pods, like an assembly line, where we installed monitors above and y- with control charts, and you could see how the work was progressing, um, through the assembly line and identify choke points. They would turn red. And, like, it was a big deal. And this is, uh, by this time it was probably early 2010s. ABA Journal wrote an article about it, and I was like, "Man, my [00:17:00] phone's never gonna, uh, stop ringing." And it was crickets. Like, law firms were not interested in standardizing and being efficient. Maybe in the scenarios, you know, like default services where you get paid per case, right? But in the big law that largely operates, you know, 80-plus percent, um, on the billable hour, that's just not a motivator. And, um, I think for us to scale with tech, we have to start doing some of that process engineering work now, um, and we don't really have the muscles For that in, in law firms. I've seen-- I can count, let's see, in the Am Law, I know, and I know probably most of them, less than 10 that ha- even have a function that, that around process engineering. Um, if there are more out there, they've managed to evade me. Um, I don't know what are your thoughts on that. Well, and you're, you're really [00:18:00] highlighting, you're really highlighting another reason, Ted, that, that applying AI in the business of law makes so much sense. Mm-hmm. Because there's much better process in the business of law. Now, a lot of these same issues of it being a bespoke service are still there, and so one of the real challenges that you have with implementing automation in the business of law is that it's just, it's just full of exceptions because what you have to do when you're in, when you're on the staff side is you're meeting the requirements the attorneys have, which come from requirements that they have from the clients. And that, that ends up being just a long list of, of, of exceptions. So- Special cases. Yeah. I mean, I'll, I'll sit down with the, with the biller at a firm, and this is someone who is responsible for putting out several hundred bills every single month. And so they'll have the six, seven, 800 bills that they've got to get out every month, and that person will [00:19:00] have a Excel spreadsheet or a Word document that is a list of every single exception that they have to deal with every single month. That is a manual thing that they have to tackle. And those exceptions are just, they're automation killers for normal automation. Like, if you just go in and say, "Oh, well, we're just, we're just gonna put process into place and everything's gonna fit into the process," well, that doesn't work. Not when you're dealing with attorneys and not when you're dealing with, with clients because the attorney's just gonna show up and say, "Well, this is a, this is a $10 million a year client. What are you telling me you're not gonna-" Do this. You're not gonna prepare the bill this way. Of course you are. You're gonna go type it up and, uh, you're gonna pull the typewriter out and put it together if that's what's required. And so yes, we are going to deliver exactly what, uh, what the client wants. So it requires an additional level of, of effort, but it's [00:20:00] still so much more realistic to do in the, in the business of law because you take that same thing and it just multiplies out like, like crazy when you get into the, into the practice of law. Yeah. You know, it's not unique to law firms to have these, um, special billing arrangements. So I used to work for Microsoft on the SQL team long time ago, 27 years ago. And, uh, my side gig, which is how this whole thing, why InfoDash exists today, I, I, um, I was roommates with a guy who worked for Maersk Sealand. They're the largest shipping company in the world by several orders of magnitude. And we're like watching a ball game, and he's like banging out spreadsheets, and I kind of look over, "What, what are you, what are you doing?" And he's like, "This is how we bill our clients." They would create these Excel templates because their creative sales guys would go out and create all these like rules to sell the-- to close the deals. So like there were about 13 [00:21:00] different variables that had to be evaluated. Um, the solution that I built for them was their stevedoring billing system. So anytime a ship would pull into port and they'd move a container either onto a truck chassis or onto a rail or put it in the storage, that represented a billing unit. And these sales guys would go in and be like, "Okay, if, if, you know, Evergreen, um, owns the ship, but it's a Maersk port and it takes this route and it's this time of day and there are these three other ship- shippers on the, on the ba- you know, on the, uh, boat, then this is the rate." Otherwise it's-- And so I-- They, they-- Their s- their ERP couldn't handle the complexity, so they hired me to come in and build a solution around it. And every time they would sell a new deal, they'd get creative. I mean, it, it kept me busy. They were clients for 14 years. Um, we [00:22:00] kind of- And law firms are no different, right? Yeah. It, it's the same thing. It's exactly that same circumstance. Same thing. And now you've got the clients imposing all kinds of rules via OCG on top of that process that's already occurring within the firm. So we've got this really complicated process in the firm that now has to match a bunch of rules. There's a really complicated process from the client that's really leveraging a third party whose goal in life is to make it so you don't get paid. Like their, their main sales pitch is you're only gonna pay 89% of, of the bills if you hire us to evaluate the bills. And so it's just, you know- Complexity on top of complexity. Yeah, that, that, that doesn't sound like a great recipe for a productive partnership. Um, when we play games like we hire vendors to minimize, uh, what I owe you, it-- that feels a bit toxic. Uh, yes. But, but AI is gonna fix it all, [00:23:00] right? We're gonna- Well, it's the reality. I mean, that's, that's, that's the bottom line, is it's the reality of things. Um, you, you talked about like AI in the '90s. Like, you know, so my first exposure was to AI was Deep Blue playing Jeopardy, you know. And then again with AlphaGo, you know, and that was like, God, that was mid-2010s. I can't remember when Deep Blue... That may have been '90s or early 2000s. But like what did, what did the state-of-the-art look like in AI in the early days when you started working with it? Well, one of the great things about, uh, about selling to West is that I, I got to-- So that put me on the management team of, of West, which, uh, w- which was very odd, selling them my little software company, uh, getting to then be on the management team of, of what was a very, you know, a very significant company at the time. And it made [00:24:00] everybody in, you know... You, you, you go into this one building up in Minneapolis, and there's 6,000, uh, people in it, and they all wanted to talk to me, not because of me, but, but just because I could talk to, uh, to the president of West. And out of that came a really wonderful benefit, which is that I got to meet and work with Forrest Rhodes, who was the creator of Keysight And Forrest had a really brilliant idea, which is that you could leverage citations, which, you know, was the basis for, for so much in Westlaw, and you could apply that to the firm's own work product. And the way that was done, it was all AI, it was all machine learning. Because how do you find citations? How do you find company names? How do you find all the information that is in documents, whether it's legal research or whether it's a firm's own work product? It's, it was all ML. And so [00:25:00] when you think about doing that work, today we take thing-- we take entity extraction just for granted, right? Because you can, you can go into Claude and just ask it a question, and it'll go do entity extraction for you. But in the late '90s, it took like actual real work to do, to do entity extraction. And, and West was, was really great at it. And so Forrest had this vision of, of we should be doing this against the firm's work product. And everybody at West was like, "Well, but if we did that, people would stop using Westlaw." And I really thought it would be the opposite. Like, if you could go into Westlaw and find not only legal research, but I'm in New York and I could find a brief that an attorney in LA wrote that was relevant to what I'm doing, I would always go to, to Westlaw in that case, which is exactly what happened. And so that, that's what West KM was. And so that's why I got to work on that was, uh, was because we sold to West, [00:26:00] and there was just some really brilliant, uh, uh, scientists working on Forrest's team and, and, uh, and other, you know, there were a lot of, lot of really smart people there at the time. So that was, uh, that was just an awesome experience. Yeah, you know, a lot of people don't recognize that AI existed long before November 2022. Um, LLMs have been around for, uh, I don't know, about 10 years now. Um, Attention Is All You Need was the famous, uh, Google DeepMind paper that really laid out the architecture for LLMs. But there are cert-certain things that, um, LLMs aren't really well positioned for, um, you know, them being probabilistic versus being deterministic. Uh, where does financial analytics land on that [00:27:00] spectrum of good for LLMs versus not? Yeah. So particularly when you're doing analysis on financial data, uh, machine learning is, is just the right-- is, is the right approach to apply because, uh, when, you know, when I'm doing financial analysis and I have certain numbers and I put it into, uh, and, and I apply AI to it, I wanna get the same numbers every time. And that's not going to happen if you're using a, a probabilistic system like an LLM. Like, the thing that makes an LLM really powerful is the randomness that is, that is inherent in how an LLM works. Well, uh, CFOs are not huge fans of, of randomness, so it's not a, it's not a great idea in terms of doing your financial analysis. So as a very final step to, to use an LLM to query prepped data that has been already [00:28:00] enriched using ML, that's one thing. Um, you know, I'm very careful with this whole concept of chat with your data because there's, there's a lot of danger with that if you haven't really done the hard work of, of properly forming that data. Like, just think about asking, what was our realization for this client? Well, are we talking about collect realization, build realization, engagement realization? Like, you have to really go through and define all those terms. And the, the firms that are doing this kind of work, they're doing all of that really hard. Like, what is it, what is it we mean when we say that? Because if you went and asked 100 attorneys in that firm, you would probably get a very wide range of, of answers. We gotta, we gotta really narrow that down. So We have our own problems with just understanding what the data is, which means that we need to make sure the data we're presenting is really solid and [00:29:00] really makes sense. And it's a great example of where LLMs are not the solution for everything. ML is still an enormously, uh, powerful technique that, that really should be used in a lot of cases. Sometimes an LLM gets used simply because it's super easy from a development standpoint. Like, if I wanna do something that really ML should be used for, yeah, I could probably develop it in a couple of hours using LLM, and it's gonna take 10 or 15 hours to do it the right way with ML. And that's why you end up with these, you know, not ideal solutions that long-term are gonna have some holes in them because anytime you use an LLM in a production environment for something that's a regular part of workflow, the additional level of QA you have to do around that, 'cause the model is constantly changing. But like, you just think about what it takes in order to make sure that you're gonna produce a level of consistent [00:30:00] results. It's a really strong argument for why you should use machine learning in a lot of applications. And for analyzing financial results, ML is the way, is the way to go. Why is it getting so little attention? Um, you know, it really seems like LLMs are, have kind of stolen the, the spotlight, and you don't hear about ML or NLP, natural language processing, nearly as much as you do generative AI. Um Do you have any sense of, of, of why we're maybe using the wrong tools when better tools exist? Well, and of course, generative AI is the combination of ML and NLP. Like, that's how you get to an LLM, right? Is you put those two things together. Uh, so, so you are using NLP and, and, uh, ML when you're using generative AI. It's, it's just that, uh, it's a very [00:31:00] different approach to get there. There's a lot of, of just core AI that is better off done with ML, even though it is just faster to get there with an LLM. So it's, it's just a trade-off between, between speed and, uh, and, and long-term consistency. Yeah. You know, I, um, I've been posting recently about a trend that I've seen where a lot of these legal technology companies, I'm taking a little bit of a pivot here, but y- this reminded me of, of something, a constraint in the system in terms of this transformation that we're, that is underway. Um, I have seen a lot of these legal AI companies snapping up legal engineering talent from these law firms that have historically... You know, [00:32:00] the legal engineer role, I feel like has been not, has been under-empowered and probably underappreciated, um, in, in law firms. And I think, and, you know, this is kind of a hot take, but I think they're one of the most important roles in the law firm, and that includes all your lawyers in your practice. I think that legal engineers, a good legal engineer is worth 10 middle-of-the-pack attorneys, um, in terms of impact to the organization and It's your ability to execute on transforming the business. So, uh, I'm curious, like, you know, from what you've seen, y-you've been around for a long time and, you know, there's been a underinvestment and I think an underappreciation from big law in these roles and, um, I don't [00:33:00] see it changing fast enough. And what's-- what feels like is happening is these big legal AI companies are hoovering up all this talent and then selling it back to the law firms for $500 an hour, two or three times what they were paying these people when, you know, they were W-2s there. So, um, that dynamic concerns me, and I don't know, man, I, I don't feel like there's, uh, the sense of urgency that I think needs to be there across the board. It, it's certainly not true in all cases. Like, there are firms that are doubling down and really making investments and fleshing out their org chart and doing the right things. So I'm not saying like this is all law firms, but I'd say more than half, um, seem to be making some really fundamental mistakes, like letting, uh, not retain-- doing everything they can to retain this talent and double down on it because they're the ones that are gonna-- they know how to-- They're where the rubber hits the road, right? They're, they're the bridge between the [00:34:00] practice and the tech, and that's a, that's a gap that needs to be bridged. I don't know. What are your thoughts on that? If I were running a law firm, I, I would wanna own those core workflows and not have that provided by a, by a third party because I think that's the real value of the firm is, is their approach and how they do that. That's the unique element is doing it differently than, than other firms. And so getting standardized because you're using the same third-party product to do that, that doesn't strike me as a great way to create differentiation in the market. So that's, uh, something else I, I like to think about and talk about is how law firms are going to differentiate themselves in the future. And today, very little, [00:35:00] very little of how law firms present themselves in the marketplace has to do with technology. Very little. In fact, I have some rough numbers because I had Claude Cowork go through the AM Law 100 list to every public-facing site and document their primary and secondary differentiator, and only eight out of 100, um, talk about technology as a primary differentiator. So 8% of the top 100 law firms, um, I think that is also going to have to change because it, it's not like the way that law firms differentiate themselves today, brand, people, footprint, practice depth, all of the things that you would-- that you think about, um, that are very traditional, uh, technology really doesn't come into the equation, but it, it, it, it is going to come into the equa- equation in the future. Maybe not in every practice area. [00:36:00] Litigation, um, you know, you being the, you know, mile deep expert in particular niche, like that's still gonna be a people- Mm-hmm ... proposition. But for, you know, the operate the company work, which represents 80 to 85% of all legal work, like technology's gonna matter. How do you see law firms differentiating themselves in the future where, where tech's gonna play a role? It's not-- I mean, off-the-shelf tools don't do it, right? Like no shade to Harvey and Ligori, but if, uh, my competitors can buy it, it's not unique to me. Um, I don't know. How do you see, how do you see this playing out? Uh, i-in the end, the differentiation is still gonna be the things that you've talked about because the largest firms are, are going to, are going to get that base foundation in place. Now moving forward, whatever time period you wanna pick, five years, 10 years from now, right? And, [00:37:00] and so it's the difference-- If we make an analogy to, uh, to another market, like your experience, um, if you go to Mayo Clinic or MD Anderson versus may-maybe your local hospital system is, is, is on their level. But, uh, in general, your, your experience of just going to a random hospital versus the best in class, like one of those two, the technology there is, is amazing. Like, if you go to MD Anderson, you're probably there because you're pretty stressed out, right? You've, you've got some things going on that, that are challenging for you, and they, they put a tracker on you so they know where you are at all times, and they have a culture around you, uh, that is, that is helpful no matter where you're at. And [00:38:00] they-- If, if you have difficulty walking, they have somebody that comes and picks you up in a cart and takes you to your next appointment. And they always know exactly where you're supposed to be, and the system is always updated, and if one appointment runs over, the next appointment is rescheduled automatically That's a base level of technology that every firm's going to have to get to a certain level in order to be competitive And, and then you're gonna have to differentiate with, with depth beyond that. So the real problem is you don't wanna be one of the firms that gets left behind because you don't, you don't make that commitment, you don't get to that base level of foundation because all your competition, all of your peers are going to, are going to get there. And then what are you going to do that's going to make [00:39:00] you great at the kind of M&A and litigation and real estate closings that everybody knows your firm for? How are you gonna add depth into that sort of base level of foundation of the technology infrastructure that's gonna be specific to your firm, that's gonna add to that service delivery on top of the foundation? That's interesting. So are you, are you thinking that there will be a certain amount of parity, um, amongst the successful-- the law firms who m- are successful making their way out of the other side of this transformation? Yeah, I, I would say there, there is today. It's just that because we are right now looking at there's this huge jump that we're expecting, we don't really pay attention to all the progress that has been made and the place that we're at today. I mean, [00:40:00] just compare where you're at in a firm right now and go back to what it was like to, to walk into a firm, whatever time period you wanna pick. But you, you're, you're running a modern document management system. You can actually find the work product, uh, in, in your firm. And imagine what life was like before having like a modern document management system. You likely have, you- you've mentioned what you do, firms that have already im- im- that, you know, have the older system with HighQ. Like firms have a way to communicate and work with their clients online Is it what we think it should be today when they're, when they're running that older system? No. But is it a big ad- big jump over where, over where it used to be? You know, absolutely. So there's a lot of parity with [00:41:00] the core technology state of where we're at. It's just that we can see today there's this next big jump that needs to occur, and then the, and the same thing's going to happen again. Y- I think that's a pretty optimistic perspective, honestly. Um, I don't, I don't think you can find y- the documents that you want i-i-in your There's been such poor data hygiene applied in law firm IT infrastructure, and not just in, on the unstructured side. Um, you know, as part of our implementation, we deploy in the client's tenant, we have to al- always touch their PMS, right? So we see every bastardization of Aderant or Elite that exists, we see it, um, and have to map to it. And, um, you know, like- And, and, and that's all I do. So it, it's not that I, I agree with you 100%, and what's the number one document type in every document management system? Whatever the- The first [00:42:00] thing on the list. Exactly. Right? So it's not that I don't agree with you, but compared to, uh, we just have everything sitting out there in a directory structure that nobody maintains at all, um, it is, it is a big advance. Is, is it what it should be? No, of course not. And, and AI-driven systems won't be either. They will be a mess also. That's, that's the thing we're missing here. Like, this isn't gonna miraculously fix everything. It will be a mess, too, because you're dealing with these huge organizations. This is the thing that, that I love and find so challenging about law firms. Like, how many places where you drive up into the parking lot, you go underground, you, you find your parking space, you go up into the building, and there are 500 people in that building that work for just one organization, and 200 of the people [00:43:00] in that building think they run the place That's the definition of a law firm. Yeah And so undoubtedly we're gonna have these kinds of, of, uh, of challenges, right? Yeah. Um, n- it's absolutely true, and I think that power dynamic is also gonna shift to a certain extent. Um, you know, like getting lawyers to even check in their documents, let alone profile them properly, right? It has been a historically a really big challenge, and like those chickens are coming home to roost now because we are finally in a place where, especially on the unstructured side, we've had, we've had ways to capitalize on properly structured data with good data hygiene for 30 years. But on the unstructured side, we really haven't until LLMs, [00:44:00] right? At least en masse we haven't. It hasn't been mainstream, the technology. I mean, the DMS vendors have not delivered anything AI related of compelling value, in my opinion. Um, and I, frankly, I think they need to get their act together pre-pretty quick, uh, because there are, there are platforms out there that are literally recrawling and indexing the entire DM corpus hundreds of millions of documents and re-enriching and r- you know, um, re-ranking for search purposes because it's so bad what, what's there. And it's-- I think, you know, I, I say that with a little bit of frustration, um, but, you know, like the, the market hasn't demanded it. Um, you know, and, and I think that's ultimately on the law firm side. We, we haven't demanded it and, and now it matters all of a sudden. Yeah. I mean, I, I live with the challenges on the billing side day in and day out, right? I'm working in 3D and Advent all the [00:45:00] time, so you know, that comment very much, uh, resonates. Resonates. But at the same time, uh, you also see like the level of complexity of the bills that are getting out the door. Like, I have a split bill with, you know, going to three different comp- three different companies that are billed at different rates, and the system's able to handle that. And so, yes, there are lots of other issues, but, uh, I can plug into that data that's sitting there and that history of the last, uh, 10 years' worth of billing data, and I can pull enough value out of that, that I can now predict when, when the client's going to pay next or what the relative creditworthiness of your clients are, and then I can use that data in order to, to, uh, prioritize what our collection efforts are. So yes, it's a problem, but AI's gonna make it better, right? So that's the financial version. You're just laying out the [00:46:00] problems with the DMS version. AI makes that better because there is a corpus to go through that you can- stick AI on and get a better result out of, and do a much better job of finding the documents that are in there, because that's one thing that, uh, that AI is really good at. Like, we've all wanted halfway decent enterprise search forever, and it turns out that the way we're gonna get there is this ridiculous process of completely ingesting all those documents with an LLM so that we have our friend that we can ask to go look for the documents for us. Like, that's, that's what's gonna-- what it's taking to finally get there. And it's, uh, it's gonna take a massive amount. It's a very expensive process. Um, very expensive process in terms of time, compute, uh, you know, it's like, it's like, uh... We, we partner with several ve-vendors in that space. I, I jokingly call enterprise search the l-land [00:47:00] of lost hopes and dreams in legal. You know, it's been, it's under-delivered for so, for so long. But, um, you know what? We're almost at- You, you should, you should do a podcast series just called that. That, that right there is a beautiful, uh- The land of lost hopes and dreams. Uh, we have a search solution as well, but it's for very highly curated, uh, data sets. Like we just did a, a webinar with Gowling. Um, we were a finalist, uh, jointly, them and us, for the Trailblazer Award with ILTA, where we took their corpus of about 25,000 legal documents and they had a million SQL rows or 800,000 that were annotations on that, and we crawled, vectorized, chunked, and stuffed into an Azure AI search index and then exposed it through our, our client extranet and it is amazing. Um, in fact, I'll include a link- Exactly ... to a, to the show, in the show notes to [00:48:00] a LinkedIn Live that we did with, uh, Al Hounsell from Gowling for those that are interested. Um, well we better- I love that and, and Al does such great work so- He's awesome ... that was a lot of fun to work with him on that. It was. Like that was his vi- he connected dots that I've never seen a client connect about our, how to use our architecture. Like, you know, that's w- w- very unique. Um, well we're almost out of time, uh, Bill, but I so much appreciate you, uh, carving out a little bit of time on your schedule to, to have the conversation. Before we wrap up, how do people find out more about you and Zebra Workz? You go to zebraworkz.com and I am bbice@zebraworkz.com, so just drop me a line. Awesome. All right. We'll include a link in the show notes and, um, thank you so much for, for joining. Been so much fun. Thanks, Ted. All right. Take care, Bill. Thanks for listening to Legal Innovation [00:49:00] Spotlight. If you found value in this chat, hit the subscribe button to be notified when we release new episodes. We'd also really appreciate it if you could take a moment to rate us and leave us a review wherever you're listening right now. Your feedback helps us provide you with top-notch content.