AI Agents Are Replacing MSP Busywork

[00:00:00] I saved over a thousand hours in the past year personally with AI. What's better is I went to my team and I shared just the opportunity and some of them came back and made what I did look small by comparison.

[00:00:15] If you want that for your team and your company to enable your team to see things and do things that were just never thought possible before and go from the hype and fatigue and buzzwords to actually building things, join us Thursday at 1 p.m. Eastern for our free build sessions. None of the hype, all of the building, all of the answers. See you there.

[00:00:38] The clients, you know, they don't care so much about the tickets. They care about the outcomes and they want faster operations. They want better visibility, better reporting, lower overhead. They want automation, security and growth. So they're looking at IT not as a function of support, but more so a function of like operational scale.

[00:00:58] Hey guys, Damien Stevens, host of MSP Mindset, founder and CEO of Servocity. Today I'm blessed to interview Robert Burke. He is the CEO of Core 12 and he's figured out how to manage 70% growth every single year.

[00:01:19] And what he just figured out is how to decouple hiring from that growth. So 70% growth, not quite as much headcount because now he's hiring agents to do real roles with real returns on investment. So if you want to know how he did that, why the industry is changing and how he became a managed intelligence provider from a managed service provider, don't miss out on our conversation today. Robert, I want to jump into the deep end with you. I want to get right into it.

[00:01:50] Let's talk about something a lot of us aren't talking about, which is real results coming from AI. And we're going to get into how you did it, all those details, but I want to talk about the growth. What kind of growth did you achieve last year? What are you on track to achieve this year? Yeah, at Core 12, we grew 62% year over year last year and we're trending 68%, just north of 68% this year. That's almost unheard of, right? That's really strong growth.

[00:02:21] So I want to back up, we're going to get into some of the how, right? But I want to back up and ask, what did you do to prepare? And more importantly, like, where do you see this going from an industry perspective? Like you were, you know, you've got a background in MSP and then an SSP, but now you're doing some different things. So let's start with why. Yeah, for sure. I think we could start with that. So my background, it was in the MSP world.

[00:02:49] I founded what eventually became Core 12 over 17 years ago. And Damien, we built a security first MSP focused on small and mid-sized companies, SMBs. Primarily architecture, engineering, construction, manufacturing, and industrial services. Over time, we realized something important, just me working with these different businesses and business owners.

[00:03:13] We realized that most clients weren't actually, like they were experiencing, weren't actually technology problems per se. Technology was just kind of where the symptoms showed up. The real issues were around operational inefficiencies or leadership gaps or financial visibility.

[00:03:35] The lack of accountability across the board with small businesses, broken workflows and bad managers, bad management systems and bad managers as well. They had people issues. They had process issues. Yes, technology as well, but technology wasn't the be-all and end-all. Really, there's like structural things that needed to be fixed. And so we fixed the infrastructure, but the business itself was still struggling, right?

[00:04:04] And so that's eventually what led to Sobo. We can talk more about that later. But as I'm looking at, as we were looking at the MSP industry shift at Core 12, and in 2025, January of 2025, conversations actually started back in 2024, late 2024.

[00:04:25] We saw that the MSP industry was, the MSP work itself was getting more commoditized and that the MSPs that are surviving and are going and moving up the stack, they're getting into advisory, business intelligence, operational consulting, now AI enablement and automation as well. So the clients, they don't care so much about the tickets, they care about the outcomes.

[00:04:54] And they want faster operations, they want better visibility, better reporting, lower overhead, they want automation, security and growth. So they're looking at IT not as a function of support, but more so a function of like operational scale.

[00:05:14] You know, the small businesses that we service, they're in like the 20 to 100, some of a couple of our co-managed clients are in the 200 range, but most of them are like that, you know, 20 to 100 size range. And, you know, I think they also see the writings on the wall, they have to get more efficient. They don't necessarily know how to get more efficient.

[00:05:34] Like the number one KPI post AI, post, let's say, mid to late 2023, the number one KPI for any business, as far as I'm concerned, is revenue per employee. And so it totally changes the game. It's not just about adding more people and doing more things. Right. So, yeah. So we looked at Core 12 and what Core 12 will be in the future state, what we want it to be.

[00:06:00] And we wanted it to be an MIP, a managed intelligence provider. And what an MIP essentially is, it's IT support, it's IT managed services. It's also cybersecurity and industry compliance like CMMC 1, 2, and 3, but also like a robust security stack, like not just adding one antivirus. Well, now you need two antiviruses with 24-7 SOC.

[00:06:26] And, you know, you need all these things like as part of your cybersecurity package, which is more like in the MSSP type of playing field. Well, if you're doing MSP, but you can't do MSSP and there's other firms that can do MSSP and MSP, well, then you probably should figure out how to do that. So we said, yes, we need the second bucket, MSSP.

[00:06:48] And then the third area of an MIP is the ability to do AI and automations. And not, when I say that, I don't mean helping people get on chat GPT. That's local AI. That's what I call it. Maybe other people call that as well.

[00:07:08] But then there's this organizational AI that are agents that are purposely built to perform a function or functions that otherwise a human would do it. Or in most cases, no one's doing it. So, yeah. And so that's the three branches of an MIP. So when, this is pretty early, it sounds like you started to realize this kind of use of organizational AI, AI enablement and automation.

[00:07:38] And then, so I guess the first question is, how'd you come to that? How'd you arrive at that? Like, what was it you saw that said, I see this trend or is it the customers or something you read or how did you get there? It really came out of my journey launching Sobo as a platform. It's a SaaS-enabled platform for companies that are looking to unpack or get clarity on what's holding their company back.

[00:08:08] Not just in technology, any area of the business, HR, finance, ops, you know, leadership, strategy, sales, marketing. So you built a new company called Sobo. And just to understand, when did that happen? Yeah. So it happened with a, it was a PowerPoint slide in early of 2023. And then we raised some friends and family around.

[00:08:37] We built the initial pilot version of the product. And, you know, we got to our numbers that we wanted to hit. And then we decided that, okay, we got, we achieved some level of product market fit. Now's the time to build it right. So we did that over the course of six months in 2020, the end of 2025, starting in June. I had to bring in a new CTO that knew what good looked like. And I had to bring in a new chief sales officer that knows what good looks like.

[00:09:08] And then we just rebuilt the product and we rebuilt the sales strategy, the motions from scratch. And we relaunched Sobo in January of this year. I'm just trying to set this up. If I understand this right, you looked at your own operations, your own enablement and automation. You look through the leadership. Yeah, I really built Sobo for myself and for my business owner friends, right?

[00:09:31] I built it for, because every business owner I know, they have two big problems that are constant, regardless of their industry or whatever. And it's lack of clarity on what's holding their company back. And it's also lack of ability to execute. You know, as an SMB, you can't, we can't hire everybody that we want to hire. And we can't just fix the problem with people. People cost a lot of money. Good people cost a lot of money. Great people are, you know, they cost even more.

[00:09:59] So, you know, so it's going to be a constant struggle, but it doesn't matter. Like, you have to figure out, how do I get access to experts? How do I result, like, figure out what my issues are? And how do I, like, structurally solve for that? And so we did that. We built that solution with Sobo. And that's what Sobo does, is it provides insights, diagnostics,

[00:10:26] for companies looking to unpack what their issues are. And it also matches them with experts in that area that can solve that problem, that have, are proven to have solved that problem many times over. So what are the lessons learned when you, you know, I want to get into something more later, but like, I just want to start with an MSP and MSSP trying to figure out, like, what is even managed intelligence?

[00:10:55] And how do we, you know, you dog fooded it yourself, right? What were some of the lessons learned there that stuck out? Yeah, so, you know, as we're going through the process of transforming into an MIP last year, I hired a, an AI automations engineer at Core 12. Well, at Sobo, but we brought him in at Core 12. And we had him start developing and deploying organizational AI at Core 12.

[00:11:27] And we, we've built out several agents that are doing the jobs that previously in past iterations of the company, I had a person doing. And so, I mean, I can tell you what we built out, we can go get into that. But essentially, yes, I went through the process of deploying organizational AI internally, so that we can learn what works, what doesn't develop the framework that we can then offer to our clients. What did you, so give me a few examples.

[00:11:56] And also, like, to me, the question is always, where do I start? Sure. Like, I could build a million things, or it sounds awesome, but like, hearing what you built will help give me a frame. But also, like, how did you even figure out what to build? Yeah, well, so first, I'll start and I'll answer that question. But I went to a conference recently and, you know, in a room of, you know, 100 business owners. And, you know, the question was asked, you know, how many people are using AI in their company? And like, everyone raised their hand.

[00:12:26] Maybe 80% raised their hand. But most businesses and business owners in that room, but, you know, across the board, they're thinking about AI wrong. They think, how do I use ChatGPT? That's not the real question. The real question is, how do I redesign my company around AI and automation? That's organizational AI.

[00:12:51] So it's all about workflows, accountability, process automation, operational visibility, AI agents that are serving a function or multiple functions, and that sort of thing. So AI shouldn't sit outside of the business or just be confined to local AI, which is what I call ChatGPT and Claude and those things. It should also become embedded into how the business operates.

[00:13:18] So the first thing that we automated at Core 12 was dispatch. So we have a Core 12 dispatch agent. It runs 24-7. It's integrated with our PSA. It's checking tickets that are coming in. It is routing them to the appropriate technician based on availability and skill set. And it's also doing follow-ups on SLAs.

[00:13:44] And we also have a kind of an extension to that that's a separate agent, but they all kind of work together. They really should work together in many cases. Sure. But we have an agent that we call Athena that runs inside of Teams because we use Teams. So it keeps the service desk updated on any failures across our client sites, any tickets that maybe they haven't responded to on the PSA.

[00:14:13] Then we'll let them know that, hey, you need to jump on that issue. It's approaching the 30-minute mark or whatever. And it's also a way for them to ask, hey, can you give me more information on this or that? So we have this kind of command agent in Teams. It could be important to Zoom or whatever. We've built it for Teams. Then we have Core 12 dispatch assigning the tickets and performing the follow-ups.

[00:14:40] And also letting the clients know via email that, hey, it's been assigned to whoever, Sam or Jonah or whoever. And they're working on it, and they'll be in touch. And if they mention the time that they want for us to connect, they'll say that in the ticket. They will attempt to reach out to you at this time. So there's logic and AI. It's not just like a copy-paste transaction. It's intelligent, right? That's the benefit of AI.

[00:15:10] It's not just automation. And so the other area that we built an agent on is billing reconciliation because, like all MSPs, we're billing, we're charging for subscriptions. We're getting billed by vendors for subscriptions. The vendor bills and the PSA don't talk to each other well. And that has to be reconciled. And it was up until three months ago, it was all manual here for the last 17 years.

[00:15:40] It's to the tune of like, it'll be 30 hours a month that you're spending reconciling the bills. So now it gets done automatic, 24-7. So let's go back. You built the dispatch first. Yeah. Why? We built dispatch first because that is actually a role that we had to constantly replace because our techs want to be techs.

[00:16:09] They don't want to do dispatch. And so dispatch is, it's also a function that is, it's repetitive and it's easily automated. It's an automation candidate. It's like, I think that's probably one of the first ones that MSPs should automate. You know, if someone wants to pick up the, yeah. If I understand it was part that it wasn't boiling the ocean. It was something that was relatively approachable, not for a first agent project.

[00:16:40] But also if I understand you, it was kind of the job you didn't have the right person for that nobody on your team wanted to do. That is correct. That is correct. Yeah. I think that's important because I think so many people think that it's just, I'm going to replace somebody. And sometimes you don't have the right person or they don't enjoy doing that repetitive work or they don't, for some reason they want to sleep. They don't want to work 24-7. Exactly.

[00:17:10] And, you know, in 10 years ago, you could get a, you could get a dispatch person willing to do the job for, you know, 40K. Now it's, you know, 60K. Well, I can spend, you know, 10 grand deploying an agent that can do that job. And, and, you know, we didn't, we didn't let anyone go. It was just, now's the time. Let's do, let's build out a dispatch agent. And we just won't fill that position when it comes back up again, because we don't need it. We can hire people.

[00:17:38] We can hire local techs for L1, L2, L3. And because that's, that's the human aspect. I mean, there's the service. There's the QAQC part that will always, as long as there's a human on the other side, there needs to be a human on your side for that. Yeah. How, so for everybody listening, watching, like, how would, how would you approach this? Would you hire a full-time AI engineer? Would you dive into this to yourself?

[00:18:07] Would you hand it off to somebody on your team? Like, how would, you know, because you started thinking through dispatch kind of hurts. The billing reconciliation, that hurts. Let me, let me figure out, you know, I think it makes a lot of sense to start with a real pain, right? You did that. But also on the pragmatic side, like, how does someone start? Yeah. No, I, yeah. So the first thing is you don't want to just throw, you know, paint at the wall and see what sticks with AI.

[00:18:34] You need to have a product requirements document drafted. Either you draft it yourself or someone else drafts it. Probably should be the business owner. If they know that role really well, they should probably draft that out. And then if you have a resource internally that knows Python, that knows, you know, N8N or what other, you know, tools that are available to build that out. You know, if you want to pull them off of what they're doing to do that, you know, have at it.

[00:19:04] With, in my situation, you know, I hired, I brought in a person, an AI engineer through Sobo. So I had a benefit. I mean, I was paying this person a hundred K salary to do work on the diagnostic engine and other things. And we just kind of redeployed them to solve for this issue and, or to build out these agents.

[00:19:26] But if I didn't have that person, then I would not try to hire a local developer to do it. I would not. I mean, me personally, I would just, I would bring in an expert that knows how to, how to deploy that agent to build it. Not only to build it, but also to, to support it, to, to DevOps it because it does need to be updated from time to time. Right. Is that what you created Sobo for?

[00:19:56] To help people find the right answers to it? That is exactly, that is exactly why I created Sobo to help businesses uncover what their issues or pain points are. Are there hopes, dreams, aspirations, what's your vision? And then, you know, matching them with someone who can execute that. So we actually have a, a diagnostic dashboard in Sobo called AIQ.

[00:20:24] And, and that was, that is essentially a dashboard where you could go in as a business owner and you can create an agent. Now, when you're creating agent, you're just kind of scoping it out. But then you're getting matched to an expert who can deploy that agent and then DevOps that agent for you on a, on an engagement basis. So, you know, you're not spending 10 grand a month to build out AI. You're spending significantly less.

[00:20:55] So I want to, I want to be clear about this. Can regular business use this or for Sobo or MSPs only or what's, who do you recommend uses Sobo? I recommend, I mean, any business, any small to midsize business owner that is looking for clarity and looking for ability to execute, they, they should tap into Sobo.

[00:21:16] So we partner with other marketplaces to, to expand our expert talent pool. We also partner with agencies. We partner with MSPs. We partner with HR companies that have an established audience base. They don't have the diagnostic engines built out.

[00:21:39] We can build that out for them, but we partner with, with partners that have an audience base and they have experts and we can make it a win, win, win for everyone. So that, that way, when you have an audience coming from, you know, your client base into Sobo, they're, they're tagged to you for life. So they're your, they're your client. They will remain your client.

[00:22:07] If you're, for example, an MSP and, you know, technology is a business area in Sobo that gets closed looped off to that MSP only. So any requests for insights or deep dives or experts, our engagements are going to go to, to that provider, to that agency that's partnered with us. But then there's also, yeah. So help me unpack this. If I understand this, I could go into the Sobo platform. I don't have to be an expert.

[00:22:34] You said build an agent, but it sounds like it's a discovery process to recruit, to basically try to figure out what the requirements are. And so like, if I understand like, why not do that? Really sounds like no risk. And then where the road meets the road is only if I, you know, get paired up, find somebody. And decide to engage. That's right. Yeah. That's right. So there's a discovery process. And that's, you know, there's no, there's no cost for that. You're scoping out an agent.

[00:23:04] You're, you're submitting a match. You'll have a conversation with an expert. And if you don't like that expert, you can swipe left. And, you know, pick the expert that, that you, you feel like you could trust to build this agent out. Keep in mind that every expert on the platform, whether it's coming through us, experts that we've personally curated or experts that we're bringing in from partner networks, they've all been vetted.

[00:23:30] These are people that have, you know, they've done HR for their, for the last, you know, 10, 15 years. Or they've done, you know, AI build out for what? The last three or four or five years. Right. They've, they've done, they built out organizational AI. So you're not trying to hire a local person that, and, you know, see how far they can go with it.

[00:23:52] Like you're hiring someone that they can take, if you don't have a products requirement document for your AI, they can help you build one out and they can go over the engagement plan with you. And if you want to engage with them, you just click a button. So. So this answers the, I don't have the talent or I don't know exactly how to do it. I don't know where to find the talent. Like I just know that I need to move in this direction. And so this is the how, right?

[00:24:17] This answers that, that some of the clarity and some of the, some of the resource in terms of who can do it. Some of the how and some of the who. I want to zoom back out for a minute and talk about why you think the industry is going this way and why be a managed intelligence provider and why you're adding like what, that's pretty new. So what does that even mean to you? Yeah. So what MIP means to me is three things.

[00:24:44] IT managed services, cyber security and industry compliance, and then AI automations. And, you know, for, for the MSP specifically, it's, you know, if you don't have the capabilities of internally deploying organizational AI, you can partner with Sobo or another agency that, you know, you'll have a rev share agreement with them and you'll make, there's wallet share there.

[00:25:13] You're not going to lose out on not only that opportunity, but future opportunities with that client. And you're, you're going to keep doing what you're really good at. You get to focus, you get to, you know, you get to focus on what you're really good at and not have to worry about whether it's the AI piece or in some cases, maybe it's the MSS piece or, or industry compliance piece.

[00:25:38] I think a lot of MSPs are already, like partnering with an MSSP or compliance partner. And then there's rev share model involved. Same with AI, like any MSP could present themselves as an MIP so long as they have all the, the ability to execute. See, we're, we're being paid to solve a problem and how we have a process for solving that problem.

[00:26:01] Whether we solve that problem internally with people that we, you know, that we employ as a W2 or whether we're using a contractor, whether we're using a partner. At the end of the day, the clients don't matter. They just, they have a problem. They want it solved. And so if they're, if you're solving their IT support, but you're not solving for their cybersecurity or their, now their AI automations, well, maybe they'll just go to someone that is.

[00:26:26] And so what we want to do is enable MSPs with Sovo, for example, we want to enable agencies that, hey, you know, this is the areas that I'm really good at. These other areas. These other areas. I don't want to spend a lot of money and time building these capabilities out, but I do want the wallet share. I want to tap into that. So we enable that through partnerships at Sovo. Gotcha.

[00:26:54] So why manage intelligence provider and why now? There's a need to evolve. I think now as an MSP. Why do you think that is now? I think it's changed. I think it's not necessarily now, but I, I would say within the next, let's call it three to five years. Okay. Okay. We have, as an MSP, you're, you're there, you're their IT partner.

[00:27:22] By extension, you're their technology partner, right? Well, they're going to lump AI into technology. So there's maybe 5% of small business, small to mid-sized companies that are intentionally deploying organizational AI. That number is going to grow. I don't think it'll grow to more than 50%. I think the other 50 that don't adopt it, their revenue per employee is going to continue to drop.

[00:27:47] And eventually they're either going to be sucked up by private equity, who's then going to operationalize at scale those AI for you. Or you're just going to continue to kind of, you know, maybe just stay in your lane and, and, and just, you know, do what you can to, to just keep the clients that, you know, aren't adopting the AI. So like you, you, you could certainly say, Hey, you know what? I'm not going to do AI.

[00:28:15] I'm just going to focus on clients that only want IT and maybe some cybersecurity, but they'll, they, they don't want to adopt organizational AI. So you could do that. For me, the clients that I want to work with are the ones that want to push the envelope. Yeah, I understand that. So is that, is that personal preference that I want to push the envelope or are you thinking this is another opportunity to be ahead and to differentiate?

[00:28:44] Yeah, I think for, um, small to mid-sized companies as private equity continues to suck up these small businesses, there's increasing competitive pressure. These private equity groups, you know, they are tapping into AI in their platforms.

[00:29:03] I'm sure most people listening saw the news, um, earlier this week on Anthropic starting a new, could be one and a half trillion dollar, um, consulting agency for AI deployment to enterprise. But some of those enterprises are those private equity groups, right? And so, you know, they're going to benefit from it because they're going to, they're going to spend the money on it as a small business.

[00:29:32] You know, if you're not adopting AI and let's say maybe not now, but over the next five years. Yeah, you're going to be in a really tough spot. Because if you're not deploying it for yourself or for your clients, then, um, you're just, you're just not going to be able to compete as well. So you, you, I agree with that.

[00:29:57] And you call it organizational AI where, um, the, the AI, the agent actually does something. And what I mean by that is it doesn't just summarize your, what you pasted. It's doing a job. It's doing dispatch in your case billing or conciliation or other things. So why does that matter in terms of how you can help your client? Well, we're, we're technology experts, right?

[00:30:27] So I think it's, um, it's like, if you're, if you're, if you're a lawyer, if you want to be a great lawyer, you got to, you got to stay up to date on, on the latest case law. And the latest, you know, uh, where, where politics is moving. Like you have to be in tune with that. So I think it's the same thing. I think, I think it's just one of those things where, um, you know, you either grow or die.

[00:30:48] So I think if I hear what you're saying, right, you, long-term you, your MSP will fall behind if you don't do something with organizational AI. And also if the customers you're serving aren't, they're probably not the growth ones. They're probably the ones the most at risk as well. Right. So it's kind of a double edged sword there, right? You're risking your business, but you're also essentially choosing to, you're, you're, you're choosing your client base as one that's most likely to be at risk or declining.

[00:31:19] That is correct. And then like, there's different strategies. We talked about them earlier. You can build it. You can build out the capabilities internally, or you can partner with someone that, that can do it for you. And then you can, you know, there's, there's a bite of the apple on that. There's rev share on, on all those engagements that happen across your client base. What do you think? Like, all I do is talk to MSPs and work with them on this and talk to them about this.

[00:31:45] This being the transformation that's going on in the industry, what you're calling organizational AI. I have an operating methodology called compounding teams where you use agentic AI and work with humans and agents. And my point is, you know, if this is happening and it is, and if it's continuing to do so, what do you think is holding most MSPs back from?

[00:32:13] Because we've, we've been like, oh, you know, MSS, like cloud computing. Let's jump into that. MSSP. Okay. Let's jump into that. But I feel like this time there's more resistance to jump in. Is that just because it's still new or do you think there's something else? Uh, I mean, it could be that it's new. That that's definitely a possibility. Um, it can also be like, look, changes, changes hard, right? It's, it's hard to change.

[00:32:43] It's hard to evolve. It takes a lot of energy, time, energy, and effort. And as business owners, like we have the least amount of time, right? We give maximum effort. So how much more effort can we give? And I think it's just more about figuring out, uh, creative ways to do it. That don't involve so much of your time. And sometimes that means stepping back and letting, you know, somebody else take care of it.

[00:33:09] So for folks that are MSPs, maybe also MSSP and they're sitting there going, you know, the MIP, you know, that sort of thing. Let's say I engage or I want to use Sobo and I want to do that. But like, where do I start? Do I start in my next TBR? Do I start with like my VC, so, or my offering or, you know, acronym superside? Do I start when I'm sitting down from a client trying to uncover issues broader than IT and then bring those back?

[00:33:39] Do I start by dogfooding it internally or something else? What do you recommend? I mean, first and foremost, I would recommend like taking a, taking a long weekend and just, you know, find some quiet space. You know, if you need to book a hotel, like spend a day or two somewhere that you normally don't and just get away from the day to day.

[00:34:04] Because it's, you can't solve your problems if you're constantly in your problems, if that makes sense. But then when you're by yourself as a business owner and you're there for a reason, like you got, you're a business owner because you're not traditionally a nonlinear thinker. Like you, you see things differently. And so you have the ability internally to, to, to figure that out. Um, then you can go to counsel.

[00:34:31] You can counsel with other business, business owners, um, you know, and, um, and then kind of flesh that out a little bit more. And, uh, and then I wouldn't go to my team until I had like a clear plan. Me personally. Yeah. Right. Yeah. I couldn't agree more. Tell me about that because I feel like that my belief is it's mindset first, then culture, then lastly tools. That's right.

[00:34:58] What's your experience been with, you've transformed a ton, right? You not only have Sobo, you're now managed intelligence provider. Like that's a pretty big shift. And you had the 17, you know, a lot like me, you've been in business for, I've been in business for 20 years. You've been in business 17. So it's, there's some identity there, uh, as a company and as individuals.

[00:35:20] So what kind of, um, challenges do you see or what kind of lessons do you see on the kind of the cultural and team side? Well, I mean, I, I agree a hundred percent with your, like your process, your framework. It's mindset first, then it's team culture, and then it's tools. A lot of times we start in reverse.

[00:35:43] It's like, we've got to get the right tools and then we've got to get the right people and then we can get our, no, it's like, okay, so you have to, yes, apply Damien's framework. Think about it that way. And, um, you know, challenge yourself, challenge yourself to, uh, to make, you know, after you go through your planning process, take a step. Uh, and I mean, you, you, you might say, Hey, you know what? I see what's happening.

[00:36:10] And from my perspective where this is goes over the next few years, okay. MSP industry splits into two different groups on group stays, uh, more support, support focused and gets increasingly commoditized. And the other group falls into like intelligence providers, operational intelligence providers. Um, and there's going to be challenges with both of those. Um, but, um, you know, you have to decide every person's situation is going to be different.

[00:36:40] Like, you know, if I'm, if I, if I wasn't 45 and I was 65, now, now's a great time to exit. Um, but if I'm in my forties or maybe even early fifties, mid fifties, you still have time in, especially if you have kids, you have things to pay for iterate. Let's iterate. Let's, let's, let's, let's make some improvements. And, uh, you know, maybe start by dispatching your or automating your dispatch or maybe build reconciliation.

[00:37:10] Like I'm sure everyone can relate to that. Oh yeah. My account manager and my account, a bookkeeper, they have to work through this at 20 hours a month. Well, that 20 hours a month times two people times what you're paying them. That's a lot of money. Yeah. A hundred percent. And it doesn't get done consistently every single month. Yeah, that's exactly right. Yeah. If you leave money on the table, you're overpaying and almost everybody is. Right.

[00:37:35] Um, and if you're not overpaying on the vendor, you're spending a lot for that really great person to throw a lot of time at. That's right. So it, it, it's a cost one way. Um, it's not always the most, it's kind of a thankless job, right? Cause it's like you always got to do something a little bit like dispatch. So that's what I, you know, the pattern, it didn't seem like you had the right person. It seemed like thankless job, somewhat jobs that are either repetitive, um, didn't have the right person.

[00:38:05] And, you know, like just where your company was, my point is you didn't come in and try to take a job that somebody loved and was doing well. Right. And I think sometimes people miss that. They go, I'm going to try to replace all of L1 support, you know? And I feel like, first of all, you're trying to boil the ocean. Like you don't know what you don't know. Yeah. Second, why would you like try to do that when you have other, you haven't figured out. That's right. Because you're, you're really good. You're really good.

[00:38:34] L2 techs were L1 techs and you're really good. L3 techs were L1 techs and L2 techs. So you need that pipeline because you're always going to need L2, L3 techs. I mean, at least for the next, I mean, I would say forever. The human is the QAQC has to be performed by that last touch. And, um. Speaking of that, on the, on the human side, have you run into the, uh, like fear of,

[00:39:02] well, if you have a dispatcher and a billing person, like is my job next? Or are you only going to need half as much? Because you're adopting things and you're moving ahead of a lot of people, right? When, when we, when we started deploying the, uh, uh, dispatch agent last year and billing reconciliation, yeah. People were saying, well, uh, I'm doing procurement. Does that mean you don't need me anymore? And we had to have conversations like multiple conversations because one's not enough.

[00:39:32] Let them know, Hey guys, listen, I hired you for a reason. And the reason is because you're really good at solving problems. And so your path to success here is L1, L2, L3. Maybe you split off into operations. Maybe you're the next operator at 412. Like there's, there's rooms to grow. We don't want to push you into an area that is, is number one, it's not your strong suit.

[00:40:00] Even if it is your strong suit, I mean, how can you be passionate? I mean, about billing reconciliation, for example. Yeah. I mean, maybe there are some people that are. I couldn't be, but I'm sure somebody must be, right? I'm sure someone is. But then you're really going to probably be good at, at other things, right? That's right. That are more valuable to the company. And going back to the number one KPI for all businesses at host AI is revenue for employees.

[00:40:28] So our, our focus over the last two years is to grow, but not grow with people. Grow with, with. So when I unpack that, before I do, I think the best thing I've heard so far is, um, what you were saying around, um, the, uh, not, not the, I want to get into the revenue per employee, but I want to go back to what you said around, like, we hired you because you're great at solving problems. Yeah. And I think sometimes it's easy for us to get stuck in there and go, well, I'm an L2. That's what I do.

[00:40:58] And the next rung has to be L3. Maybe you're good at operations. Maybe you're great in another area. That's right. So your job changes, or even if you're that dispatch, maybe you're just so good at solving problems. I could really use you in this area. Uh, I know personally, like not that exact one, but a similar one. I needed them to move out of dispatch into another area and solve a more valuable problem, but not until I fixed the dispatch issue, if you will. That is correct. Yeah. So I love that.

[00:41:26] It's really still thinking about the people and the, really the best and highest use. Um, that's right. I want to contrast that with, cause some people get really nervous and maybe misread like revenue per employee. Cause if that's now the North star metric, we're tracking that too, or we're making big improvements there. Um, like that is great. That makes the business more valuable, right? That sort of thing.

[00:41:51] Um, how do you think about revenue per employee now in this kind of with their post AI world? Well, yeah, I mean, to me, um, the way I think about it is that it's, it's good for everyone. Like it is, it should be the true North star because it's good for everyone.

[00:42:13] It sets the tone and then you figure out, okay, if my revenue per employees, let's say 200 K, we needs to be at 500 K. How do we get there? It gives you like, that's why they call it North star. It gives you, it gives you a direction to margin. And, um, I think it's, it's important because I mean, look, we're, if you're, if you're privately held, uh, which probably a lot of people listening to your podcasts are, uh, you're not owned by, um, private equity.

[00:42:43] Um, you know, you have to compete against them and that's their North star. So help me understand, like you're going to grow, let's say you're with 30% or 70%, right? Or a hundred percent or whatever you're going to grow. Does that mean you'll not hire anyone? Does that mean revenue per employee being a higher one? You're going to have, you're going to be more effective. So you'll still hire. It'll just be less. Oh no, no. We'll still hire.

[00:43:09] I mean, we have, um, we have a, uh, pipeline we developed with a local university here in town, uh, Georgia tech. And so we, um, we, we bring on to, uh, L one interns in their junior, senior year. And we, uh, give them full-time summer and then they do 10 to 15 hours a week in the fall and the spring. And then, you know, through graduation and they get a guaranteed job offer when they graduate.

[00:43:38] So we always have, cause you, you know, we have to develop that pipeline. We have to protect that pipeline. Um, and so, yeah, we'll get just a constant stream of L ones. L ones moving into L twos, L twos moving into L threes. I think there's many MSPs that are, are, have already looked at, um, you know, let me just hire some really good L ones and L twos and then have a retainer model for the L three, L four type work. I mean, there's no, there's no harm.

[00:44:08] There's no foul on that. So long as you find the right provider and that can increase your, your revenue for an employee. Um, cause you know, hiring a really good L three, L four, whatever you want to call it, like a really good project engineer, you know, you're paying 150, 200 K. Yeah. That position. So. So is your thoughts on revenue per employee. It doesn't matter what the number is 1 million to 2 million or 5 million to 10 or whatever. Like if you're growing, then.

[00:44:38] You know, I'm making up the number, but if you were at 50 people at five and you're going to a hundred people at 10, which would be a horrible revenue per employee, but still in my little fiction math, right. Does this mean, well, you're still going to add, but maybe instead of a hundred at 10, you're going to need 85 or 75. That's right. And it's, it's a, it's a lot easier culture to manage. Like if I can do 5 million with 10 people versus 30 people, a lot easier to manage. Yeah. Yeah.

[00:45:08] Yeah. Now, do you think this generally accrues to the owner solely, or do you think this means that you develop these super high valuable kind of 10 X players, which means that the, the pay per person goes up as well? Yeah. I think, yeah. Increment the pay is going to go up as well. Right. But you're, you're paying more for those employees per person.

[00:45:37] And like 10 employees at, you know, doing, doing 5 million a year, you're going to be able to pay them more and people need to make more at the end of the day. There's also fewer people. There's fewer good people. I could, with the boomers aging out and birth gap. I mean, you know, we have some real structural issues, not just in our companies, but like, you know, we're, we're all experiencing the societal dysfunction right now. So we have to figure out ways to, to navigate it.

[00:46:05] And, and yeah, so ultimately I do think it will, it will drive wages up. I think that's another benefit to that. I'm just glad to hear you say that loud. Cause so many of I've talked to about this and, you know, they're saying that they believe that, but, um, you know, so many people just read articles of like, oh, there'll be less people or like they'll, they're looking for the negative.

[00:46:26] And I think there's going to be some companies that will take shortcuts and they'll have negatives, but I think there'll be some really amazing opportunities that just weren't there before, um, for the true problem solvers. Um, so kind of any like last thoughts or wisdom, having been through this journey for any owners and MSPs that are sitting there going, what do I do? You know, how do I take the next step or make this transition?

[00:46:56] Like you've been through, you've got the battle scars. Yeah. Yeah. Um, well, again, I mean, I would, I would take some time to yourself to, uh, to do, to do some planning, do some, you know, do some praying, um, figure out, you know, what, uh, where's your company now? Where do you want it to be in two years, five years, 10 years?

[00:47:20] And, um, not take out any outside noise, like, you know, really, truly give it thought and not get distracted by noise, but other people, you know, telling me things, um, that they're probably not, um, um, qualified to tell you, right? So, um, yeah.

[00:47:42] So, I mean, that's the first thing I would do is like get off, get off the technology for a day or two and, um, and do, and do some soul searching. And, um, you know, if, if you're an MSP and you're trying to figure out how to evolve beyond traditional support and build a more strategic business, if, if anything I mentioned earlier was, um, was, was interesting to you, I'd love to hear from you personally.

[00:48:09] So, speaking of that, Robert, for folks that would like to reach out to you, pick your brain, talk about what you've been through, where you're going with Courtois or Sobo, what's the best way to connect with you or find you? Yeah. Best way to connect with me is just my email. It's robert at Sobo.ai. S-O-B-O dot A-I. Okay. Don't make, uh, don't miss out on the opportunity to reach out to Robert.

[00:48:36] Um, a wealth of, uh, wisdom and hard fought lessons and, uh, you know, more than we even have time to get into today. So don't miss out on that opportunity to contact Robert. Um, Robert, it's been a blessing having you on today. Thank you so much for being here. Likewise, man. I appreciate it. Thank you for the time being here. Yeah. It's good to be here, David. Yeah. I appreciate you. Yes, sir.