Most boutique founders know AI is a tidal wave, but they keep trying to bolt it onto a billable-hour model. In this episode, member Mike Gibbs explains how he decided the old leverage model was structurally dead for knowledge work and rebuilt his firm around AI-delivered expertise. He walks through converting consulting know-how into productized tools, shifting pricing to SaaS-like licensing, and protecting proprietary IP by running private, controlled models instead of public systems.
What you’ll get from this session:
- The trigger that forced a full business-model reset in a GRC firm
- How to turn “consultants on a whiteboard” into an always-on AI application
- Pricing mechanics: replacing hourly billing with license, maintenance, and user-scale economics
- How to prevent IP commoditization using private models and locked-down training
- A practical starting point: solve one annoying problem, then build a focused AI strategy to avoid “shiny object” chaos
Why it matters:
- AI reduces delivery time, making hourly pricing structurally fragile
- Productizing expertise decouples growth from hiring and utilization
- Private-model deployment keeps tribal knowledge proprietary while still scalable
- Firms that wait for proof will be displaced by firms that rebuild first
TRANSCRIPT
Greg Alexander: Hey everybody, this is Greg Alexander. You’re listening to the Pro Serv Podcast, brought to you by Collective 54. If you’re new to this show, this show is created for founders of boutique professional services firms that are trying to do three things. Make more money, make scaling easier, and make an exit achievable. And on this show, we’re gonna talk to a member, Mike Gibbs, who is in the process of reinventing his firm to be relevant in the AI era. And he wrote a very interesting article that caught my attention, and thus my… I issued a request for him to be in the call of the podcast today, excuse me, and that… the title of that article is From Billable Hours to Billions of Parameters. And I wanted to have Mike on the call today to talk to you about how he’s reinventing his firm in the AI era. Mike, for those that don’t know who you are, would you please provide an introduction?
Mike Gibbs: Yeah, sure. So, thanks for having me, Greg, and, I’m Mike Gibbs. I’m the CEO and founder of SureStep Risk Management, and now Onyx AI Labs. And we are a company that has had a long history in the governance, risk, and compliance space, providing consulting, advisory, implementation services, and so on. So, I’ve been doing that for about 15 years in 7 different countries now at this point. So, yeah, been around for a bit. That’s who we are, what we do.
Greg Alexander: Alright, so let’s start at the beginning of this transformation, the trigger, if you will. So, you built your firm the traditional way, billable hours, smart people, leverage model, and it worked for many, many years. Then all of a sudden, you decided to change. So, what made you decide to disrupt your own model with the new AI tools?
Mike Gibbs: So, I think… I mean, look, we’ve been all hearing about AI for quite some time now. I kind of started this journey about a year ago. It was around January of 2025 when I started I guess maybe taking this whole AI thing seriously, and you start to realize this isn’t a fad, this is a tidal wave that’s coming, right? It’s not a small ripple that we’re all gonna, you know, it’s not some kind of, like, new eyeglassware or something like that that we’re all gonna look at. This is… this is something that’s changing everything, and it’s gonna run over everything that’s in front of it. If so, you kind of have a choice to make. You say, alright, either we’re gonna surf this wave and ride it out, or you’re gonna drown, right? And… the writing on the wall was pretty apparent early, like, I mean, if you look at, you know, what AIs are good at, you know, it’s, frankly, it’s that kind of knowledge worker stuff, right? You know, so legal, accounting, consulting, those kinds of things, and that’s right in its wheelhouse. And, you know, what do you have as a boutique firm as an advantage over an AI, right? Well, it’s all that stored knowledge that you’ve managed to put and gain over the years, and your experience, and so on. So, how do you take that and make that wrapper-friendly for the AI world? And that was really the task I put about myself. And, that’s how we kind of got started on it. It was just kind of this realization that, like, if we don’t change a year from now, we might not have a business. So that’s where we were.
Greg Alexander: Yeah, okay. In the article, you talk about, from labor leverage to AI leverage. And you discussed how the traditional professional services model, which was hiring really smart people, and then charging them out at a multiple, and that’s how we earn the profit. But you’re suggesting that that model’s going away, and that now the leverage is going to be in the machine, so to speak. So, would you expand upon that for the audience, please?
Mike Gibbs: Yeah, sure. I think the term I used in the article was kind of going from just a carbon-based delivery, being, you know, humans and biological delivery, to almost a silicon or a silicon-carbon hybrid, right? And I really think that’s the way it’s going to go. So, you know, I can give you a bit of an example, like the, you know, like, we’re in the midst of building out some new applications that really do take our consulting knowledge, like, the stuff that I normally would have had some of my consultants go and sit in front of a client with a whiteboard and whatever. We’re taking that, we’re putting that into an AI, so we’re fine-tuning AIs, essentially, to learn our consulting knowledge, and how we train things, and how we would have spoken to a customer, the knowledge that we would have passed on to them. And then what we do is we put that into an application. Now the client can essentially access that anytime they want, anywhere in the world, however they want to get it on whatever topic, right? So, we’re taking our knowledge, and rather than us being the mouthpiece that’s giving it there, we’re putting that into a silicon shell, right? Or an AI shell, and saying, here, now you can talk to this guy. It’s not as if I’m buying you a beer across the table, but you’ve got all the same knowledge and all the same capability, and they’re available to you all the time at a far less overall rate than you would have if you were billing for us individually, right? So, that’s what we’ve done.
Greg Alexander: So that takes me to the third question, which is the fear-based question. So, when people hear this, and when people like you or me talk to them about, you know, these types of things. They get really worried about pricing, and they say, you know, when technology reduces delivery time. Clients expect lower prices. How do… how did you deal with that?
Mike Gibbs: Boy, there’s a lot of things we had to deal with on this. You know, it was the pricing thing, you know, people worried about their jobs, clients worried about their jobs, right? We work in the compliance space, right? That’s a really… it’s an interesting one to be in, because most people don’t really… you’re not really adding to the bottom line, in the eyes of most of the higher-ups in a bank, for example, right? You’re a cost center. And if they say, geez, if I’m using AI to do this, you know, what’s going to happen to my role? You know, and then for my people, it was like, well, if I’m going to get a project done in, like, 10 minutes, well, then how are we gonna charge for hundreds of thousands of dollars to pay my salary, right? So, you know, and the answer to that was, like, you know, forget your billable hours. We turned that into an AI-powered software. That somebody can use and access, and we charge for that. Right? And rather than charging, you know, maybe we’re only charging effectively, like, you know, a dollar an hour for your time, but if we’re multiplying that over, like, 500 users times, you know, 25 or 100 different companies, well, then all of a sudden, we’re actually making more. Right? And this is what software companies have been doing for years, right? I mean, like, you know, I mean, ultimately, I mean, they… they have a software piece that’ll grab a piece of data, do something to it, and give it back, or… you know, give some consulting advice at the back end of it or something. It doesn’t matter what it is, you know, I mean, we’re taking away and we’re multiplying the effectiveness of communication and effectiveness of the knowledge that we can disperse.
Greg Alexander: So, so you used to charge by the hour, now you don’t. So what is the pricing?
Mike Gibbs: Yeah, so, it… so right now, it’s kind of almost going to… kind of a SaaS, almost, pricing model, right? Like a software as a service. You know, and people can buy our stuff however they want to do it. We’ve got it air-gapped because some of our clients need that kind of regulatory security. So people can buy it as a chunk, right? So we have some clients that are probably going to buy it and say, hey, here’s a half a million bucks, have a nice day. All right, cool. We’ll charge maintenance on it, the same way as the software companies have for years, and it’s been a pretty good model for most of them. Seems like that’s probably the right way to go at the moment, anyways. That might change.
Greg Alexander: I think it is. There will be additional things. Outcomes. Well… just moving away from the billable hour to, you know, some recurring revenue model. SaaS-like model makes a ton of sense for you. You know, when I speak to members, sometimes they say to me, I can’t do this, my clients won’t let me do this. And I chuckle at this, because here you are, somebody who operates in probably the most highly regulated world there is. And you’re doing it. So what do you say to those people who are so naive and they’re making an excuse by saying, I can’t do this because my clients won’t let me.
Mike Gibbs: I would say… you’re probably gonna need to go find new clients. I mean, you know, look, that’s a reality of it, right? I mean, like, look, there’s some people out there, some clients. Look, we work with banks literally on all sides of the planet, right? Some of them are absolutely not going to do this. They are going to want to talk to a person, depending on where you are geographically, what your industry is, and what your clientele’s like. Lawyers are famously reticent about moving off of their old way of doing things, but the ones that are leading the charge, the most successful law firms I know are ones that are embracing this, right? We’ve got our colleague, you know, John Arnott, of course, you know, who’s absolutely fantastic in terms of his message and what he’s doing in terms of, like, helping old fuddy-duddy companies get on board with AI and really make things better. You know, look, once you demo something to a client, you say, look. I’ve just helped you create a control that is absolutely validated against the regulation. You… here’s your defense memo that you can send to the regulator and say, hey, we’re on side with this, and you’re covered, and here’s all the reports you need to send to your stakeholders, and it’s all done automatically for you, simply because you said, hey, how do I deal with this new rule in a chat interface? And you get all of that? It is pretty hard to argue the efficiency, right? And especially in larger organizations, they understand that AI is coming, and, like, our mandate, like, from one of our clients is, like, hey, we’ve got to find a way to figure out, like, a 30% efficiency gain in human time over the course of the next 24 months. They’re only going to do that with AI, they know they’re only going to do that with AI, and there’s simply no other way. And if our tool is going to be a part of doing that, well, then, you know, we’ve had some of my people on my team were like, well, we’re going to be taking away jobs from people, and yada yada yada. Yes, we will, and that sucks, and it’s terrible, and it’s brutal, but the world is changing. And, you know, as I ended off my blog with you guys, it was like, I mean, like, look, you can either be on the leading edge of this, you can either go and build this yourself. If you don’t, somebody else is going to. And you’re just gonna get run over and left behind. So, you know, yeah, this is a sea change in the world, right? And you’re either gonna be on the list… you’re either gonna survive and move on and adapt and change, your clients will do the same, or they won’t. And, you know, I chose that, hey, you know what, I got a mortgage and kids and dogs and cats I gotta take care of. I’m not content to just sit here and be consigned to the dustbin of history, right? So, that’s where it came down to. So, that’s a long-winded way of saying, hey, you need to… you might need to educate your clients first.
Greg Alexander: When I read your article, because I’ve gotten to know you, you’re a super nice guy, and you were being very, very polite with that, but you did address it. I’m known for being quite a bit more blunt, so my answer to that question is kill or be killed.
Mike Gibbs: Yeah. I mean, that’s just… I mean, this is the world we’re in.
Greg Alexander: Alright, the other thing I wanted to ask you about. What you do is a tremendous amount of intellectual capital that you have built up over the years. The hard way, you know, each implementation, one at a time, all over the world. Like, you are a true expert, and you have a tremendous amount of tribal knowledge. Now, all of a sudden now, you’re coding that, and you’re putting that into AI, did you have a fear? That, as a result of codifying it and putting it into AI, it becomes a commodity.
Mike Gibbs: Well, that’s only if you’re silly enough to put that into a public model. We run all that stuff internally. Like, we have our own models that we’ve downloaded, we run, and we’re training internally, right? So when somebody buys our software, or they’re getting it as a SaaS, it comes with a model embedded already. We can change the model, we can train the model, our clients are not allowed to train the model, they’re not allowed to touch that thing, right? So there is that fear out there, and of course, like, that’s part of what we do, right, is AI governance is part of what we do as a rule. So we were maybe a bit more fortunate and a little less naive than others, maybe, in terms of you know, what you can say to a public model and, you know, what you shouldn’t. So I think we’ve been very good about keeping that domain. So there is that misconception that, hey, if I use AI to do something, I mean, it’s all going to get out there, but you can run models privately. You know, there’s, you know, you can do it a bunch of different ways. You can do it on Google, you can do it in Azure, you can do it however you want, or you can just do it on your local desktop, right? But yeah, keeping that knowledge to yourself, I mean, yeah, I mean, you don’t want to let that get out into the wild too much, right? You know, no, we don’t have nuclear codes or anything, but it’s certainly not stuff I’d want to let out into the general domain.
Greg Alexander: Yeah. Yeah, I mean, we built an agent, as you know, it’s called CoFounder, and it sits on top of multiple models. And we have the same fear that you do, which is, you know, I’ve got 30 years of intellectual capital in my head, right, and I don’t want to all of a sudden, you know, get commoditized. So, yeah, I mean… and it is a misconception. I just don’t think people realize that you can run these models in a private way that doesn’t…
Mike Gibbs: Well, lucky for C54 members, that’s the part of my next blog that I’m writing is all about that, so… so there we go.
Greg Alexander: What, where did you get started? You know… actually, let me… let me back up. Our members right now are overwhelmed, and they’re like. AI, AI, AI, where do I get started? Do I start in sales? Do I start in delivery? Do I start in internal firm operations? Like, what do I do? And, you know, our point of view, Mike, as you know, is we say one use case at a time, and really, you’ve got to start with what’s the use case. You got started immediately, it sounds like, but correct me if I’m wrong, in what you offer clients. Like, you converted your service into an AI-enabled service. Is that correct?
Mike Gibbs: Well, there was a step before that that I think is important, right? I… like everybody else, I mean, I’m not Superman, I was kind of overwhelmed with all this AI stuff as well, right? And, for me, an area that was driving me nuts, was… my kids, as you know, are athletic, they’re in very high-end sports and everything, and… the accuracy of the tools and the stuff we were getting for analyzing our kids’ games was just terrible. It was really bad, like, stuff from Huddle and Instat and these other companies, and it was just awful stuff, and very inaccurate, and… it was driving me nuts. Okay, there’s got to be an AI that can do this. And I started just pounding around and looking at things, and… yeah, I found… I just started making my own tool, which, ironically, I’ve got a company who wants to buy now, which is hilarious. But, but it’s… I would suggest, like, look, there’s two things that I’ve found, so after making a few wrong turns as we’ve gone down this, a few things I’ve learned. Figure out a problem that’s annoying you and go do it, because that’ll drive your curiosity and drive you to kind of go learn about it, right? It doesn’t even need to be a business thing. Mine wasn’t a business use case, it was like, I’m just… I hate going to the rink and somebody’s saying, yeah, your kid didn’t get a… he only got two assists last week. I’m like, what are you talking about? I got 3. You know, just trying to answer goofy questions like that. That was enough to drive me to at least get into it and get involved and start learning about it. In the business use case side of things, where’s your biggest itch? And I would suggest it’s, like, usually where are you spending the most man hours? That’s the one that’s usually the biggest problem. And, you know, and I think, you know, the third part, I would say, is to just start somewhere, you know? When you get a little bit further into it, it’s very easy to go nuts once you realize the power you have in your hands to create answers to all of your problems and all of your business issues. When you get to that point, after you’ve solved that business thing, and you’ve actually been able to prove to yourself that, hey, you can do this, and you can solve a business problem with this. I think it’s very important to take a step back, and I didn’t. This cost me probably 3 months of time. I went and just tried to solve all of our problems all at once. Put together a strategy. Figure out what do you need to do, and you know, I talked to you about this before the call and before the recording, Greg, like, we’ve put together a strategy saying, alright, what are we as a company? What’s our core values? What are the things that we need to pull that off? And as I told you, we have, like, a three-pillar strategy that we’re using going forward, right? That we’ve designed. These are the three things that we need to have in order for our values, our knowledge, and our capabilities to have that value proposition put out into the world. That’s… these… these three pillars are what we’ve got to build on. You need to do that. And even if it’s just for internal stuff, if it’s just for your own internal operations and so on, pick those foundational pieces for you to build on, and those are going to give you a guiding light, because I’m going to tell you. All the rest of you are all founders as well, you’re all probably like me. You go, hey, shiny, and you immediately take a look at it, and boy, I tell you, the allure of that is great, because you don’t just have to look at something and say, oh, that was kind of neat. You can go build it, right? And you can build it in days. So the allure of that shiny, especially for people who I would submit probably most of us in this group are like that. You know, we’re problem solvers, right? We want to go fix a problem. Oh, good, I can do that one. I mean, I have had, at one point, I had, like, 9 different Cloud Code windows up on my screen, building different things, and I’m like, okay, this is probably pushing things a bit too far, I really need to focus. So… once you get that strategy around how you want to go forward, things get a lot easier. You’ve got to create that… that guiding… that guiding star for you to follow. Otherwise it’s very easy to get distracted. Very easy. So… I hope that kind of helps give a little bit of guidance in terms of which way to go.
Greg Alexander: Yeah, yeah, it does. You know, when we did it. My fear was, and I wasted a lot of money, is that I actually needed software engineers to do this for me. And I hired a development shop and spent, sadly, a few hundred thousand dollars in many months and got crap. And then Claude issues this new model, and me, who’s a non-software engineer, learns about vibe coding.
Mike Gibbs: Yep.
Greg Alexander: And now, you know, the average person can actually produce these things. It’s just… it’s just stunning. So the advice that I would give people is just, there’s a lot of, what’s the politically correct term? Snake oil salesman out there telling you that you can’t do it, and the reality is you can’t do it, so just get comfortable with the tools. So let me… I guess maybe I’ll conclude with this, and we’ll save the rest for the private member Q&A. Now, all of a sudden, you have like, your addressable market just exploded.
Mike Gibbs: Yes.
Greg Alexander: Because the barriers, you know, for most professional services that were on the billable hour was, you know, every time you sign a new client, you gotta go hire more people, and finding people that are good, recruiting them, negotiating their pay package, developing them, retaining them. It just was always the obstacle, and now that’s no longer the obstacle, so has your, like, level of ambition expanded?
Mike Gibbs: Oh, dramatically. Dramatically, yeah. I mean, I’m not, you know, like I mentioned to you before, I mean, I’ve, you know, the, you know, one of the pillars that we’re building on our AI governance tool, I mean, like, we’re… we expect that that could be a ubiquitous tool that every company on the planet uses.
Greg Alexander: Damn.
Mike Gibbs: You know, I’ve got a few pieces that I expect are going to be very big in the world. You know, and again, I’ll be happy to share… Patience, I’m about a month and a half away from being able to show that stuff off to everybody, but I mean, I’m getting close, but… but the reality is, is, you know, yeah, and like, for my people, like, again, it’s still… and I gotta… I gotta really point out, too, like, it’s still a huge adjustment for my team, right? I went to them, you know, last January, and I said, guys. Start thinking about AI. I went to them again in June, a little less gently, and said, hey guys, giddy up, right? You know, I need everybody to start thinking about things, and really how they’re gonna do this. And actually, like, literally just last week, I delivered a tool that says, hey, here’s how you’re currently doing your technical implementation, and the way you’re doing it sucks, and it’s taking way too long, and we need to do this all differently now. I’ve gone and given them a tool that actually does, like, 90% of the work for them. So now they’re all of a sudden going, -oh, right? And they need that wake-up call, you know? And more than… and I say that it’s almost like I need to give them a kick. It’s not so much I need to give them a kick in the butt, there’s just so many people that don’t understand this world or the impacts. So I’m trying to get the people on my team that are willing to buy in. I’m trying to give them all the education and the training and the patience I can to bring them into this world. Understand, stop thinking about how you’re gonna do this. How are you going to get an agent or an AI tool to do this? Right? Change your thinking. We can… we can make an agent do 90% of what you’re doing, right? But I need your knowledge to direct it. It doesn’t make you obsolete. Your knowledge is still what we need to make this agent, but for us to grow as an organization, I can’t be worried about you getting a cold. I can’t be worried about you not being able to show up for work because you got a kid getting married or something. And, like, those are all cool things for humans, but the reality is, the world we’re going into doesn’t care about human needs. It just doesn’t. And, you know, and we just saw the previous video here, you know, a gentleman asked a very pertinent question, what about the people thing? The people are still needed. Like, I mean, we still need to talk to people at the high level, right? We still need to guide this stuff and manage it. But you gotta think about, like, hey, you’re not the delivery mechanism anymore. Right? It’s now a silicon delivery, or a carbon-silicon hybrid delivery, right? It’s not a… it’s not a… it’s not the people anymore. And so, working at it, it’s… it’s difficult, but we’re getting there.
Greg Alexander: Yeah, good. Well, listen, I get asked this question all the time, and they say, hey, Greg, you know, what’s an example of somebody doing this? And you are, you are an example, and there’s not enough examples right now. So when I… when I saw your article, and I learned a little bit more about what it is you were doing, I’m like, damn, I gotta get Mike on the podcast, because people, you know, sometimes folks need proof, you know, before they make the leap. You’re rare in the sense that you didn’t require proof, you just had a hunch and you went with it. So I admire that about you, but just know that you’re acting as a role model here, and I greatly appreciate it. On behalf of the members, thanks for sharing your example with us. Please keep us posted, you know, because you’re just getting started. And how this evolves.
Mike Gibbs: Well, the next, blog will be out. I’m hoping to put it out in a couple of weeks, in a bit in advance of when I’m supposed to have it do, so, the next step in the journey will be there, and like I said, if anybody has questions, just reach out to me through C54, I’m happy to help any of the members here any way I can, so…
Greg Alexander: Awesome. Well, thanks, Mike. I appreciate it very much.
Mike Gibbs: No worries, thanks for having me on.
Greg Alexander: Yeah, let me conclude with, just a couple quick, quick, calls to action. So, if you’re a member, and you’re looking for this example of an Era 3 firm, Mike is that example, one of a few these days. Look for the meeting invitation for the private member Q&A, and we’ll have a detailed conversation with Mike about what he’s doing. If you’re not a member, and you’re listening to this podcast, you know, on your podcast player, and after listening to this, you might want to become one. Go to Collective54.com, fill out an application, and we’ll get in contact with you. But, until next time, I appreciate your attention, working me into your busy day, and I wish you the best of luck as you try to grow, scale, and someday exit your firm.