Small Law Firms Finally Have the Edge, If They Don't Blow It

Collin McKee

Welcome back to the Endeavors AI Podcast. My guest today and I figured out on our first call that we've quietly followed each other up and down the East Coast for about 30 years, whether it be DC, Philly, West Palm, and now we're practically neighbors on the Gulf Coast of Florida. Keith Conrad runs Verbatim Legal, and he works with small and medium-sized law firms to modernize their infrastructure and carefully implement AI without blowing themselves up operationally. Keith, it's good to have you, sir.

Keith Conrad

Thank you for having me on. I appreciate it, Collin. It does seem we've been close, but not close. Look at our career paths.

Collin McKee

Excellent. Give the audience, if you would, the one or two sentence version of what you do and who you serve.

Keith Conrad

We focus right now on small to mid-size law firms, as you said, to help them take advantage of some of the technologies that are out there, and certainly AI, which gives them access to the same type of legal technology that only the larger firms and government agencies had up until now, with their on-prem and very expensive systems. As these become available, it's nice that they have them, but the question is how do you use them? What's the best way to implement it? How do you implement it without bringing down your firm? There's a tendency to want to do everything at once or to do nothing at all. So we try to work with them, come in, demystify it and provide a plan, and if necessary, we actually get involved in doing the program managing for them.

Collin McKee

Great. You mentioned in our first call that you draw a clear line between the practice of law and the business of law. Walk our viewers through how small firms historically competed with big firms and how that's changed.

Keith Conrad

Well, it's changed in that basically they never did compete. They really could not compete. They didn't have the manpower, the money or the technology in many ways. That's not to say that some didn't try, but it was sort of all or nothing. What has happened, I think, is that the business of law has become much more prominent, and the practice of law is an entirely different focus. These firms need to focus on the practice of law, but the business of law can't be ignored. It never should have been, but it had been. How they go from A to Z, what their processes are, what technologies they're using, where their training levels are. It's one thing to be able to compete with Big Law, but it's another thing to have the infrastructure and the internal integrity to do that competition, and you've got to have that. If you try to take on more doing it the same way you've always done it, it is going to fail. The technology isn't going to save you. So it does come down to understanding the business of law: how they get paid, how they organize their cases, all the details of the business of law, and making sure that dovetails carefully into the practice of law.

Collin McKee

Turning back to you personally, you went back to law school when you were 50. What did that teach you that a pure tech consultant would have missed?

Keith Conrad

Law is interesting. A lot of it has been around for a hundred years, and some of it hasn't been around for a hundred days. Credibility at the table is really what you have to have, and you need to be able to understand why law does things the way it does, why things take a while, what has to be done, the issues of privileged information and data and what we do with it. In order to really have any credibility when you go to a firm that's been around for 20 years and say "you should listen to me," you've got to have some street credibility. Working with bigger firms scares smaller firms, and if you work with smaller firms, bigger firms sort of write you off a bit. So it was important to me to understand and get back to the basics and be able to tell them, "Look, I know. I've been there. I don't practice, but I've been there, and we can talk the same language." For me it was more like I had to go learn Chinese because that's what they were speaking.

Collin McKee

Excellent. Good plan. In your opinion, where's the real agility advantage for a 10-lawyer firm that a 500-lawyer firm can't really replicate?

Keith Conrad

It is the agility factor. These people that run the firm, management and staff, can get together and quickly make a decision. They can turn on a dime, and from a legal perspective, that's light speed. The larger firms have to get the partners together, and there are all the other pieces that come in, the junior partners, all kinds of levers, and they basically get to where they need to go by throwing money and people at whatever direction they want to go. Smaller firms, if they have the internal ability, can turn on a dime if something comes up. One of the things we talk about a lot is that in the last 15 years there are probably 15 to 20 practice areas that didn't exist before: cyberbullying, cybersecurity, all sorts of things that these small firms can take advantage of, and do it very quickly, and if they're set up properly, be very effective. But they have to be as good or better than the bigger firms in order to be competitive. There's a lot of opportunity out there, far more for small firms today in my opinion than for large firms, because even large firms are jobbing out some of their legal work to smaller firms if they're able to handle it.

Collin McKee

Yep. And you mentioned on our first call that it used to be that the larger firms had access to these large language models and technologies that the smaller firms didn't have. Now, thanks to technology and AI, the playing field has been leveled. Now the smaller firms have the advantage because that moat isn't around the large firms anymore, and their agility is the differentiator. They have the flexibility to make these changes with the tools that can give them leverage.

Keith Conrad

Exactly right. With the old on-prem systems, you were hard-coded and geared to the big money-producing practices, and if you didn't fit that mold, you had to fold into it, change your taxonomy. There were global settings. It took quite a while to go through the old change control processes and get everything signed off on, which is not flexible, for the larger firms. This does in fact give smaller firms the ability to have cloud-based legal technology, expand through that, increase or decrease as needed, and really have the same exact tools at a much, much lower cost. That's a key factor.

Collin McKee

Agreed. Plenty of things from our first call stuck with me, but another one is how you are advising people to use AI as opposed to what the masses are doing with it: just using it as a shinier Google search, just as a tool. You mentioned that AI is less like buying a faster computer or tool and more like hiring an employee. That's a wonderful way of looking at it. If you could unpack that for our listeners, I'd really appreciate it.

Keith Conrad

For years we've looked at technology upgrades, bigger, faster printers, faster computers, what have you, as the mainstay of what was considered upgrading or updating your firm. Those tools took the existing processes and made them better. AI, properly done, and I'm preaching to the choir here, is something that's going to actually do the work. It's like anyone else you bring into a firm. You want to make sure that it's trained properly. You want to make sure that the various process steps are carefully defined. And the most important thing, I think, is you have to keep an eye on it. You have to be able to watch it and check on it, and those processes have to be in place. It's very similar to bringing in a new employee, far more so than technology. The technology, at least from a business side, is not the key here. It's the prompting and understanding the exact nature of what it's doing, making sure it's right, and then making sure it does it correctly. Those steps are sometimes difficult to implement because people want to get moving quickly. But you're hiring, and if you look at it that way and develop your steps that way, you're going to have a much smoother integration with AI.

Collin McKee

Yep. So there are two training programs involved: the training of those using the tools, and the training of the tool itself. There's a lot of work to put into it for it to work correctly. A lot of people want to look at it as unboxing a Christmas morning present that's just going to work once you put some batteries in it, and that's a dangerous way of looking at it.

Keith Conrad

I think that's the old model: you get a faster computer, they download the image onto the computer, and away you go.

Collin McKee

Right.

Keith Conrad

That's not the way this is going to work, and it's certainly not the way you want it to work in a heavily regulated environment like law. There are some horror stories. You need to understand how to do it, and that's the niche we're filling, not just with AI but with the business itself, because before AI can do it, you need to make sure your actual manual business processes are functioning, with no gaps and no duplications. That's really the cornerstone of what we do. It's essential that you look at AI as adding to your firm, not replacing a computer.

Collin McKee

Yep. And you even go as far as telling some of your clients to remove AI before deploying it. It sounds like that's the exact reason: you want to make sure all the foundational processes are well established before we start throwing fuel on the fire.

Keith Conrad

Well, we do a pretty complex current state analysis of the firm, and we tell them that we won't do business with a firm that doesn't agree to that, because we need to know if they're currently using AI in any capacity. Have they done a comprehensive background? Do they understand what AI is doing? Do they understand that if you had a gap in your process and you didn't account for it, AI is going to amplify it? Who wrote the prompts? Who's doing this? It can affect your whole business, and potentially expose you to all the various penalties involved. So we start with this comprehensive current state so that we can understand everything about that business, probably more than anyone has ever done. It turns out to be one of the most important documents in this business, or quite frankly any small business: starting, ending, what if this, then that. All of that has to be accounted for, and that's generally not available. It's all sort of fly-by-night, and that's a very dangerous thing for AI.

Collin McKee

Yeah. Before you map out a plan to get somewhere, you need to know where you're starting from.

Keith Conrad

Exactly. There are a lot of simple analogies. You wouldn't do this on a road trip, so why are you doing it here?

Collin McKee

So when you do that current state analysis, what pops up as the blind spots on a pretty regular clip?

Keith Conrad

The tendency, particularly with small firms but not exclusively, is that you've done it a certain way for a very long time, and maybe there are some automated pieces using different technologies, but they tend to be very siloed, to use that expression. From an operational standpoint, you've just done it so long you don't recognize it. There are manual steps in these automated processes. "You have to take it down the hall and put it on Bob's desk." Each of those becomes a possible problem, and they don't see it. When they talk about their automated case delivery, well, it's automated here, then there's a stop for a manual piece, then there's an approval that's manual, then it goes. We need to identify all of those. Not everything can or should be automated, but we need to know where the manual pieces are that you've been comfortable with for 10 years, because that could really throw a wrench into what you're trying to do long term.

Collin McKee

So how much would you say is technical versus a cultural "we've always done it this way" sort of problem?

Keith Conrad

I'd say they're about equal, but for different reasons. Lawyers are very risk-averse people. "So what's the big improvement?" You need to provide them with defensible reasons to do this at all, and dispel the stuff they've seen in magazines or online about how wonderful AI is, that it can do this and it can do that, and get that sort of stereotype out of their heads. The technical piece tends to come from thinking it's a one-stop thing that can fix this right away, and it probably can, but you also have to take a look at the whole picture. I think I've said before that if people, particularly at small firms, continue to practice law the way they do, with the old practices and the old technologies, it's very likely that in five years they really will not be around. They'll either be working for a more advanced firm or finding another line of work. It's like not putting in a telephone years ago. You have to do it. And the other thing we bring is that we'll do it for you. It's not a question of you having to figure it out. We're happy to do it for you, and that's the way they should go, whether it's us or anyone else.

Collin McKee

So without being a fear-monger, how do you bring up the potential pitfalls, the possibility of sanctions, with these firms when it comes to the use of AI?

Keith Conrad

Most of them have some sort of knowledge from something they've read, some horror story out of some firm, and it's becoming more and more prevalent. I think there's a desire to do it the right way. The question is how do we do it without shutting down the firm and starting from scratch? It's not so much fear-mongering. It is an essential piece of moving from a small firm taking cases the way you used to, to actually getting into the game of what's going on in legal today and what's going to go on in legal. It's not fear-mongering that you're going to put your firm out of business. It's that you're going to be behind. If you think it's difficult to compete now, it's going to get even more so. Not to mention the other side, Collin: there's a whole new generation of clients and young lawyers that have been on a cell phone since they were two years old. There's an expectation out there, and some firms meet it and some don't. Usually, by providing them with defensible alternatives, "this is why we're doing it, this is the right way, we know more about this than you, you should listen to us," we get past that. I always said that working with law firms is the best thing I've ever done, because you have people that are being paid by the hour and argue for a living, and that's a tall order. I learned very early in my career that was really the best thing. You come prepared. These people bill out at $1,500 an hour. Do not waste their time. I think that's a rule of thumb at every law firm: be prepared, understand what you're talking about, and be able to defend it.

Collin McKee

Great segue. So you have a firm that says, "I want to deploy AI." What's your actual sequence? Where do you force them to slow down? What's your process?

Keith Conrad

I think the first big fear with AI seems to be people losing their jobs. "If I do all of this, I've had people with me for this period of time." So the first thing we've done is develop our own set of places for these people to go within an AI environment. Generally speaking, they're not going to do their own prompting; we're going to send that to professional firms. But it's important that they know how to prompt, and it's something they can learn, because, as I mentioned earlier, the output has to be looked at and addressed carefully, and you need to be able to do that. They need to be able to communicate. It's not that they're going to sit at their desks and develop, maybe they will, but it's an offline thing. They need to be comfortable with how to talk to AI and the various things that are available, so they can be proactive and say, "Look, Perplexity has a new piece out and we think we could use it." I think that's the job security for people who want it: to be engaged in the back end of AI. And the law firm still has to run. So the first thing we do is say this does not mean you need to get rid of anybody, for the most part, that you don't want to get rid of, and we'll find a place for them and show them where they're going to be. Once we get past that hurdle, there's the expectation of "we can just put this in here and start moving." They understand they need to do certain things, but they don't understand that when those things happen, they have to have lawyers and infrastructure to take care of it.

Keith Conrad

So it's pulling it back and saying, "Yes, you can do these things, and here's a schedule. Here's what we need to look at, and you need to get past certain points." For example, can you provide oversight to AI? Do you have the staffing and the ability to do that? If you don't, then we start very small, and at some point they can bring it back into the firm. This is probably off topic, Collin, I'm sorry, but the one thing they need to understand is that the large models they're going to use are public. Right now, the focused, legal-specific ones are the purview of large law firms and law schools. So they're going to have to be very careful about what they enter into most of the AI systems that are cost-effective at this point. Putting privileged information in is a death knell. You don't want to offer that. So that's another skill set. "I realize what you want to do, but in order to accomplish that, you're going to have to go into your client data. We need to build in the safeguards to make sure it doesn't wind up in a training routine for ChatGPT or whatever we're going to use." The other thing is making sure they understand that this is a task-oriented introduction right now, not replacing the legal function. That's the thing to try to get their arms around, because they've always got people on staff who say, "I've used it, I can do it, watch this," and that's great, but is it in the firm's best interest? Personal usage is one thing. When you're dealing with people's sensitive information, you've got to build your policy and build the prompts to make sure of that. That's the most important piece.

Collin McKee

Yeah, we've heard that here a few times. There are policies in place at different firms: if you're going to use anything, it needs to be the professional, paid-for account. It can't be going into the public ChatGPT. That's really important, and it's a good reminder. Switching to the aligned advisory model that you built: you told me that you don't bill hourly with your clients, and your compensation is tied to measurable outcomes. Why do you structure your payment that way?

Keith Conrad

Coming out of the DC legal market, consultants charge a lot of money. I've got a lot of experience that is valuable, and yet these small firms cannot begin to pay the hourly rates we were charging in Washington, and the job is similar. So we work on a performance basis. If we're as good as we say we are, you're going to experience a decrease in your expenses, an increase in your performance, and better marketing, and we're willing to take a piece of that. Again, my belief is small firms are the future. They really are. We want to associate with these small firms now and help them grow. In some locations, like Washington, we can take some interest in the company as payment for that over the long term. In Florida you can't do that, so we take bonuses based on measurable results established by ABA metrics, and we measure the improvement that we claim we can bring. The other side of that coin for us: one of the biggest problems of being a consultant in DC was, first of all, you're always to blame for anything, and secondly, there's an immediate conflict between us believing we need 30 hours to do something and a client not wanting to pay for 30 hours. You might wind up somehow breaking the process, making it less effective. In this situation, we don't make money if they don't. So we're not going to make bad recommendations or put in some software that isn't the best they can have. It's less of an argument. We get into it, but we won't recommend it, and if they choose not to do it, we'll leave. We have a contract out, because it's not going to work if you're not willing to step out there. We're certainly willing to work out whatever plan makes sense for you cost-wise, but you have to do this, or there's no point in us being here. So it's looking eyeball to eyeball with them. We're partners in a sense, and we have very direct conversations about that, which has been very helpful. It cuts through a lot of nonsense, gets us to where we need to be, and gets them to stop worrying about us so they can go back to worrying about their clients. We're going to make money, so of course we're there.

Collin McKee

I love that point. It's almost counterintuitive that they'd take issue with you billing the same way they do, but the way you do it aligns your incentives in the exact same direction as theirs.

Keith Conrad

Exactly. That's changing too, though, I'll be honest. Firms are getting away from the billable hour. A lot of firms are not going to do that anymore, and they do hybrids, because again, the clients have cell phones and can do things remotely, and coming into the office for a one-hour meeting is not where it is. It's 15-minute increments, and a lot of changes operationally. If you can demonstrate, "This is how you were doing it, and this is the money. Now here's the potential income if you change that model." So again, it comes down to being knowledgeable and able to defend what you tell them.

Collin McKee

Yep. You've also mentioned that you prefer vendors to handle the implementation piece, with your role being more of the blueprint and oversight. That separation of strategy and execution is pretty important. Most firms need both, just at the right time. You come in, lay the foundation, build the blueprint, and then allow those who focus on implementation to do that piece. I think that's a really strong strategy as well. So looking out three to five years, where does AI create a real advantage for small firms, and where does it create a real liability?

Keith Conrad

I think the advantage is that it allows them to get where they want to go faster than they normally could. If an opportunity is there, they can take advantage of it faster than a lot of the larger firms. It also gives them a chance to establish a focus in a particular practice area that might normally have taken many years. There are a lot of ways to be aggressive with AI and make things faster and smoother. The other thing we bring to the table is that there are a hundred ways to do it, and we have the benefit of having spent a lot of time in a lot of firms. "For this particular problem, a lot of firms have it, and these are the three ways they've solved it, but for you we're suggesting this." We save a lot of time by being able to reference other experiences. They're not a Silicon Valley firm. They're a law firm. Just because something can happen technically doesn't mean that it should, or that we should do it now. So we're sort of the voice of reason on some of that, and it gives them an advantage, because they're going to focus on where they want to go, not everywhere they could go, which is all possible through a strategic approach to law.

Collin McKee

Yep. So if a managing partner is listening to this, and you want to give them one piece of advice, something they should do this quarter, what do you think that should be?

Keith Conrad

I think it's for every firm, I don't care what the size is. They need to get a sense of where they are. Do a current state assessment, and do it with someone like us, or not, but don't bring in your cousin who's really good with computers to do it. You need someone who understands the business of law. Why is this overlap necessary? To the untrained person, it might seem like just overlap, so let's get rid of it. No, it's important to your docketing and other things. So they have to do that, and they have to put the time in. There are other firms out there that will do this, but as I said, we don't even start until we've got that done, and it takes as long as it takes, because they're not paying us by the hour. If we need six weeks to get it done, that's what we're going to take. If a firm wants to make any changes, in my mind, at this point: get an understanding of how your firm works, A to Z, today, so you can always refer back to it. Because I guarantee you, Collin, and again I'm preaching to the choir, three years from now we're going to be talking about something else, and that blueprint of how the business operates is going to be as valuable then as it is now, because along the way we'll update it and make the changes. It needs to be there, because if you forget something and you automate something manual that doesn't work, you're just going to make it worse, faster. Once we get past that, it's generally a pretty impressive document. Some keep it online, but they can print it if they want to and refer to it for a number of reasons.

Keith Conrad

And I'll be quick about this one, but the other thing we find is that we come in and ask, "Why are you doing it that way?" "Oh, that's how we did it when I got here." "Okay, and how long ago was that?" "Eight years." Government goes back and refers to things by presidential administration: "Well, that's how Calvin Coolidge taught us to do it." Now we've got to take a look. So when you hire people, the training is automated, the process is known and followed, and you can determine if there's a deviation. There are a lot of benefits from a management standpoint, but also from a growth standpoint.

Collin McKee

Couldn't agree more. You need to know exactly where your current state is to be able to map out where you want to go. So the simple takeaway is that AI is definitely not magic. It needs to be governed. It needs clear processes, and then you can start to benefit from all the wonderful leverage it brings you. Keith, this was exactly the good, bad and ugly kind of conversation I like to have about AI. To our listeners, and we'll leave it in the show notes, where can they find you?

Keith Conrad

We're at verbatimlegal.com, and we're building a better presence on Facebook and social media. We're growing, so that's always a painful process in that sense. But best to go online and set up a time to talk. We don't charge to chat with people. If you've got a firm, maybe you need us, maybe you don't. It doesn't do us any good to come in if you don't, and you need to be comfortable with us if you do. It saves us time, and it's better for them. So just give us a call. We'll chat and figure it out.

Collin McKee

Awesome. I appreciate the conversation, Keith. For managing partners listening: get out there and get your current state analysis. It will do you nothing but good. Subscribe, and we'll see you on the next show.