Moving From AI Novelty to Operating System: Maturity Curves for Mid Sized Businesses

01when we were doing when when generative AI hit the street and it was a big splash two years ago and and you know it's January of ⁓ twenty four and ⁓ everybody was talking about chat GPT. It just became a thing and every company was like well what do I do about it and and then there wasn't even like a lot of information on the street about you know ⁓ what to do about it. What what what should you have a policy? And everybody started saying I needed a policy. I feel like That opening line for every one of them is Welcome everybody. I'm Veer. That's Jim. We're ⁓ the functional executives. We're here to talk to you today a little bit about ⁓ mid market companies, the work that we do. We're talking about AI, we're talking about hype and how maybe you don't need to buy into the hype side of it, but that this is something important. Maybe you want to pay attention. Yeah, absolutely. And we're also gonna talk, I think, a little bit about fear. And it's pretty pervasive out there and and not just ⁓ I'm gonna lose my job fear.

02⁓ different kinds of of fear that are motivating people around this whole AI trend. I think that'll be a good conversation that we'll have too. And you know, just some of the mistakes that you might make along the way. And you know, I tell my kids all the time, it's okay to make mistakes. That's how you learn. You just don't want to make huge, massive mistakes. So hopefully we can help give you some ideas on how to ⁓ avoid the the big ones, but make the little ones because that's what that's how you learn. All right. So if you don't know us, I'm Veer. I've spent my career doing technology and CIO work and now private equity operating partner type work, ⁓ helping companies implement strategy and align their IT strategy with their either the investment thesis or their purchase thesis. ⁓ and ⁓ and I'm also an entrepreneur here in Nashville, Tennessee, ⁓ doing a healthcare based startup, ⁓ serving ⁓ the community of folks who are struggling with and dealing with dementia.

03Yeah, and ⁓ Jim here, two time PE CFO, ⁓ twenty-five years in finance and an entrepreneur myself. I do a lot of the same work that Veer kind of outlined for mid-sized companies, private equity based companies, ⁓ primarily. And then I also co run a five unit, soon to be five unit location, ⁓ brand with my wife. It's franchise for dog hygiene and wellness here in ⁓ the Denver Metro. And so I'm actually using a lot of what we're talking about. I'm not just talking about it. I'm like applying it in my business real time. So which I I definitely feel a lot more comfortable about doing that because if I mess up, I'm messing my stuff up, not somebody else's. Right. And so hopefully you're gonna see and you're gonna hear from us both from a pragmatic standpoint of guys who are on the field or playing the game, but also we're out here doing work with companies helping them ⁓ do ⁓ bigger long-range strategic planning. So you're gonna get both sides of that and ⁓ hopefully you walk away with some ⁓ nuggets of things that you can both apply ⁓ in your day-to-day.

04⁓ but also take back and talk to your teams about Yeah, I did want to ask you, just kind of kick us off if that's all right. IT for a long time has talked about maturity curves. Mm-hmm. You know, where are you in some process? Whatever that is, you know, it I think it was, you know, ERP has one. you could you could pick it. And now we have an AI maturity curve. Where Do most companies sit right now in terms of AI? Way on like the the low end of the maturity curve. And I think the like so if we talk about maturity curve and and try not to try to make it something that's objective, ⁓ you take standards. ⁓ on the IT side, we have things like ITL or COBIT. These are industry-wide standards. There, there's a rubric that you can apply against different factors like How well on a scale of one to five do you do customer support? How do you do security? How do you do project management? How do you handle, ⁓ you know, how's your data cleanliness? And ⁓ you take those scores and you add them up.

05And now you have something that's a little bit more objective about like how mature are you? And that's a structure that we can apply across a whole enterprise, right? We can look at it and say, How mature are we with our processes in operations and marketing and finance? ⁓ and that gives us maybe a better picture of well, when we say a maturity curve, right? It's really more about how mature are you as a business? Do you have standardized processes? Do you audit those processes? Do you ⁓ adhere to them? Do you hold yourself accountable to, you know, did you did you do what you ⁓ said that you were gonna do? So AI maturity. Right, could be measured maybe on this idea of like, we've got, hey, ⁓ we you know, somebody's got a co-pilot license or it's baked into our ⁓ Microsoft stack, all the way up to, you know, are you using ⁓ some type of chat bot? To are you do you have processes enough that you can have ⁓ coaching for your call center being driven by ⁓ AI ⁓ tools to Are you doing clinical in in the hospitals decision support, right?

06Are you feeding information into a system helping ⁓ doctors and ⁓ nurses diagnose based on patient symptoms and ⁓ telemetry coming off of the the the patient? Just to summarize, are you saying that most mid sized companies actually sit like at level one? Because I think mid-sized companies are this is not a a hate conversation. This is like they are a product of their ⁓ of their own success, right? What what has grown many of them was not adherence to a lot of standards, ⁓ or you know, it was that they had a moment, they had that they had a sales focus, they grew really big, right? The revenue is high. ⁓ but often I think Jim, you and I both have walked into a lot of organizations. We've talked to a lot of organizations where when you, you know, pull the cover back, you're like, ⁓ you know, this is a little bit of like you're doing well in spite of yourself. ⁓ good for you. I mean, like that's the power of, you know, that's the you know, that's that's that's sheer human will holding a lot of things together sometimes.

07Yeah, no, I I completely agree with that. It's ⁓ it's really well said. I will also say I think so we've we both agree that most of these mid-sized companies are at a level one. They're just, you know, emails being rewritten by AI, you know, the cliche thing that everybody says now to kind of like turn their nose up to people that don't use AI as an operating system. Leadership though. I think leadership thinks they're like at a level three or a level four at a lot of these places. I think they've asked for a project or two and ⁓ and and then they've been told, yeah, we did it. Or, you know, the the fact that maybe somebody gets told that they've got a chat bot and they're like, ⁓ that's sophisticated. And you're like, well, in the grand scheme of things, it's it's really not. I mean, it c it can be, but most of the time it isn't.

08⁓ but yeah, I think there is a little bit I it when you say that about like leadership would say that they're higher up on that scale, I I'm I might even qualify it to say I would think if you asked some of the leadership, they would say that their people are further mature because they you know that that they're that they're ready for and and using it when they're really not. I you're right. I don't think they understand best practices and how you should really use the technology. As a matter of fact, I did start, I forgot to tell you about this, I did start looking at the what is it, the Claude Code Developer Certification? Okay. Did you see that? I didn't So I started looking at it and so I took the forty page, you know, here's what you need to know to take the it's a proctored exam. So I think it's a legit I think it'll be one of the first legit certs. Yeah, exactly. So I started so I dropped the forty page PDF into Claude and I said, Hey, well why don't you why don't you walk me through this? I didn't do that.

09I didn't d not do that poorly on the ⁓ the first section, which is or the first section that we chose, which is like I wanna say thirty, forty percent of the test. Those are the type of skills that people don't have right now. And there I think the fact that there's a certification shows that people are starting to apply real rigor around this AI skill set and guarantee most mid sized companies don't have Somebody who's certified in Claude Code. Yeah. But it is an interesting takeaway from last week because I think there were a lot of people over the last year doing, ⁓ a generative AI certification. ⁓ Microsoft is digging in deep. I just had a gentleman of a b buddy I know here in town who got his AI with SQL Server, which is a brand new cert that Microsoft put out and he's one of the first ones that's got one of them and ⁓ that I think what it says is that the bigger companies are starting to actually invest in rather than, you know, LinkedIn training, which is great for general knowledge.

10But here you and I talked about how ⁓ anthropic seems to have really kind of gotten the the that that first mover advantage into the into the marketplace with a lot of the you know corporate ⁓ use cases, right? I think it's like the last couple of like numbers have been like seventy, eighty percent of companies are using some form of anthropics tools. Yeah. Yeah, they've they've overtaken in the enterprise space, they're everywhere. I think ChatGPT at what nine hundred million users. And not daily active users, I don't think, but nine hundred million users, I think they still have kind of the consumer market. Everybody knows the name. ⁓ lockdown. But it's the enterprise. The enterprise seems to really be leaning heavy into Claude. Yeah. And anthropic, I should say. That's probably a better way of saying it, because they have a lot of different products now. got these certifications, we've got ⁓ ways to start showing value.

11⁓ and I think what we're kind of working towards is the fact that not only do we need to make sure that our people are well trained, but that you know we were talking maturity, that the maturity of our organization is there so that we can take advantage of AI to ⁓ you know to improve something. Yeah, I mean it I I do think it needs to add value, but I think the first step, my my personal take is the first step is everybody's got to get past AI as a novelty. And I think that's where we are in a lot of places right now. I mean, like you mentioned chatbots. So many people are using these chatbots, but they don't I mean, if you go online and try to use some of these smaller company chatbots, they do nothing. You get one prompt in and then it's like, let me get you a human. You know, so I mean, if there was a chat bot that could take you end to end, that'd be a little bit different. But I I count all those one prompt and done chat bots as ⁓ we have something we could show the board that we did. And ⁓ that's that's kind of novelty in my opinion.

12You know, the user sees something new that it's almost a a marketing play. What I do think is literacy and getting people I I say AI as an operating system all the time, but AI by default. So a long time ago in a land far away, actually not too long ago, there was ⁓ there was a guy named Paul Bogue who wrote ⁓ some pretty good books on transforming your your business digitally. And this was way back in kind of the early two thousands, mid two thousands. ⁓ Two thousand tens, I should say. His phrase was digital by default. And I used to say digital by default over and over and over again. I kind of tried to make it my own. And this was back when I started, there was one pizza company that I worked for. I don't think it's a bad thing, I've just Marcos. I worked at Marcos for a long time. When they were really growing, it was seven hundred and fifty locations to or two hundred and fifty locations to nine hundred and fifty locations was kind of the ride that I was on while I was there. ⁓ so really rapid growth.

13But it was a small company and just barely getting their feet under them in terms of back office. Now the operations, the food, everything that they did was great. But the back office was still we were living in a world where that we had probably seventy percent of the people that still like to use paper. I mean, if you remember back to that time, I mean, websites were like you couldn't do everything online that you can now. So if you go back in time and you look at, hey, we've got this split of people, we used to send every week we would get pizza boxes taped up. Inside the pizza box was From every location, every corporate store that we had, the receipts and then the nightly ⁓ reporting signed off by the manager that they ticked and took. No serious yeah. And they FedExed them once a week. And so the receptionist would come in and we'd have like fifteen, twenty pizza boxes stacked up on top of each other. And then accounting would go through, pop those open, and that's how they got the receipts. And then of course, you know, we went to scanners and phones and things like that.

14But we were trying to push everybody from Yeah. I'm sorry. I'm sorry. Keep going. Keep going. This old paper way of doing It's fine. It is. It's fun. I mean, it's kind of funny looking back. I mean, it was really smart. I mean, they used what they had. They had to have the receipts and they wanted to make sure that they had controls in place. So it was a good control for the time. It was a little lagging. Like you would hope that they would I mean, 'cause computers have been around for a while by then. But you know, like there was an investment in scanning technology and all that that was needed. And then you have phones, and so now people can just take pictures and then we started to do Very quickly, very quickly, we moved electronically. But that's how it was done. And it was a shift though, because we had people that had already been working at the company for fifteen, ten, fifteen years. That's how they did it. They weren't necessarily tech savvy. So the same thing right now is like we were trying to move everybody to digital by default.

15Well now I think you you almost have to start moving to Who are used to it that way. It's like an AI by default. That's my that's my entire point, is like we have to work it into every part of the organization. We have to find all those pizza boxes that we have now and we have to scan our receipts and send them in. I mean, I so okay. Like what you're describing just fourteen years ago, I think was the tipping point where ⁓ the conversation went mobile first. Twenty twelve ish, you know, like like that like there there were enough iPhones ⁓ you know, that were out there, the Blackberry had started to like lose, you know, like like the like like ⁓ come come off of like, you know, that everybody's moving to a slab phone. ⁓ yeah, we had these moments, mobile first, digital first. I I agree. I think we're having a, you know, are are are you ⁓ you know, if if it's not AI first, you know, it's you know, AI at the front of the front of the pack. Excel, email, websites, all of that stuff is invisible now.

16Fifteen, twenty years ago, it wasn't invisible. It was front and center. It was ⁓ we need these new skills, we need these certifications for this. That's where we are right now with AI, is we're trying to not make it so visible. It needs to like start to fade into the background and just become a way of work. Huh. Well ⁓ I I didn't think I'd stump you on that. I just I don't think that I just and the other thing is about that is I don't think it's the AI flashy project. Like companies that go after these flashy AI projects are the ones that aren't going to really get the most valuable i th or value. The ones that are gonna get the most value are the ones that do what I just described, in my opinion, which is it starts to become such a way of doing things and people are so proficient at using it, that's when they'll start to see real value and then they'll be able to do some flashy things. That's so I think it's like, you know, don't don't put the cart before the horse.

17You really gotta train your people and get them comfortable with it and make it their daily operating system. And they need a good example. Like I'm and I think that's the like of ⁓ I think you and I, we've we've spent time the last couple of months working with several clients of ours, ⁓ educating, right? I mean, the the conversations really were are have been education conversations and and not, you know, I think both of us have done a little bit of like, you know, kind of small product demos to people to show them what it could look like. So that they could then go, I can do something like that. Like like like the the dots don't connect very easily for folks, you know, when they when you tell them you can automate something, find one workflow that you do daily, weekly, and ⁓ decide that that's not something that needs your hands to touch ⁓ physically every week. Yeah. And this isn't like, you know, this audience that that we're talking about, this isn't the YouTube, you know, I've got a whole open claw set up in my basement with a stack of Mac minis.

18This is like folks that are pressed for time and learning this stuff is I mean, it it's it's coming along a little bit slower in in some pockets of the market. And that's who we're really trying to talk to here. I I would be curious if If folks would let us know probably in the comments, if you're watching this on YouTube, you know, what it is, what was the unlock for you? And I'll ask you that question. Like, what was the unlock? Like when you saw AI do I mean, you were all hyped up over Mannis for a while. Because it impressed me ⁓ f with AI, the unlock or the the hype up over that was that its ability to not just answer questions, its ability to finish. It is like and and it's still like I'm I'm I'm working with other tools maybe more today, but I I I use it, I used it earlier today. It does a great job ⁓ being given a goal. And that's something even in the past week, something that ⁓ ⁓ open AI and Codex has been taking on. I know Claude has something, right?

19Go there's a there's a slash goal skill and and and it's kind of written in a way to kind of push the the AI to finish, right? I I think that's what when when you and I have to ⁓ when we work together, I think you know you you used to use ⁓ you ⁓ use the term ⁓ wrestle it to the ground. Mm-hmm. ⁓ yeah. Yeah, I still use that cliche. Thank you for thank you for making fun of me. Thank you. Well no, I I think it's a great example Is it's just business jargon and we're trying to stay away from that and I use it, dang it. For some people though, I don't think they really like like you had an intent behind what you said when you said that to people. And ⁓ and I think, you know, you you and I have talked about how different people have missed the intent behind it, right? That that wasn't a I I want you to own it. It was a I want you to finish it. I want Yeah, all the way. Like when you get to the point where you're like you hit a blocker, I want you to figure out what you have to do to get past that and keep going.

20I don't want you to just stop and say, I got blocked. And that's kind of where we're at. Like that's that that is some of the more impressive pieces with some of the AI right now is and MANIS was one of the first kind of agentic tools where the agent took agency for itself and would see a problem through, would hit blockers, would try to run around them, you know. ⁓ I need to do this. I don't have this tool. I'll go find the tool. I downloaded the tool. ⁓ it didn't work. I'm gonna try something different. It re like you could watch its chain of thought and ⁓ and it was impressive. 'Cause you know, it was like a dog with a bone. Like it just was not gonna give up until it got to the end. Right, right. So most companies are kind of at level one. Here's some ideas on how we can, you know one of the big ideas on how we get to a little bit further down the road is just to make sure people are using it and it becomes possibly a bit more invisible as just technology as as just another part of our life and our way of doing work.

21But there's also a lot of issues that you can run into, ⁓ especially as you get new people that are really excited. If you you know, you're gonna have one or two people on your team that are super excited about this. And if they are, they can make some mistakes. There can be some real there can be some pitfall. As a matter of fact, we had, you know, ⁓ you know, we've been doing this podcast for a while, and now we're starting to get some comments. People are letting us know, hey, you're not thinking this through. So I mean, I think, you know, frustrating as some of those comments might seem to be, we just kind of take them on the chin and go, actually, not bad. We should address that more. And there are some pitfalls. Absolutely. I think you know, I've spent my career in in in some regulated industry space. I mean you have two and and so the rules, safety, precautions, risk management, all that's important.

22Not trying to throw that out and be like, you know, it's w the Wild West, which I think ⁓ you and I have have said, you know, w one of our maybe, you know larger frustrations as we talk with friends and and and folks is ⁓ people being reductive into things that are and making everything kind of this binary. It's either good or bad, it's either right or wrong, you know. And so if you look at this problem and say there's only one answer and it's fear and ⁓ and and to combat the fear, you need lots of risk reduction. And for risk reduction, you need lots of rules and And ⁓ you need a lot of guard, you know, ⁓ guardrails and and ⁓ and policies and ⁓ I think I we were talking about this the other day. I made the joke that it's ⁓ it's Robocop two, where they gave him all the rules and he had like 4,000 rules to to follow and he couldn't enforce the law. ⁓ but but okay. I mean, you know, if you if you if you see the joke all the way through to that end, I mean you're like, well, you know, even sci-fi says that that doesn't work. Yeah, I mean anything taken to the extreme.

23So the the middle, the uncomfortable middle, right, is is what I think we are im asking folks to try to embrace a little bit more deeply is those people, those of you that listen to us and that are senior operators, you you have a a lifetime's worth of gut experience. You've seen where things where you can make some measured risks and ⁓ And and you need to embrace those and ⁓ and and and kind of find the middle ground here because the opportunity, if you will, empower people with these tools is ⁓ is is really, you know, you know, extraordinary. Yeah, no, I I completely agree with that. I think I still struggle with where the middle ground is though too, because there's a lot of partnering with because everybody who's curious can do what they want to do, there's a lot of partnering. I think partnering becomes even more important right now. And part of that means, well, did I talk to finance? Is this actually drive the business forward or is it just cool? I mean, should I spend the next forty-eight hours of my life or of the business's time working on this?

24And then if I do, did I talk to IT? Did I talk to marketing? You know, I mean if it's customer facing, did I talk to marketing? But assume it's not customer facing, did I talk to IT? Is this I mean I can build it now, but is it exposed? 'Cause nobody in accounting or marketing is building SOC two compliant Claude code projects. You know what I mean? Yeah. But you know, the the the the balance there is your Excel spreadsheet isn't either. And so I think the thing here is that the balance is starting to do the right level of comparison, right? I mean like i because if we if we take it and put it on the seesaw and make it a yes, no left, right, up down kind of a Well, that's true. conversation. And even the middle, if we say that there's three, well, that's still just as bad, right? I mean the fact is is early when we were doing when when ⁓ generative AI hit the street and it was a big splash two years ago and and you know it's January of ⁓ twenty four and ⁓ everybody was talking about Chat GPT. It just became a thing.

25And every company was like, well what do I do about it? And and and there wasn't even like a lot of information on the street about you know ⁓ what to do about it what what what should you have a policy and everybody started saying I needed a policy I and I was at one company where ⁓ I helped write the first policy which was a reactive policy that they needed but the point of the policy was ⁓ we already have a policy. We have a policy it was right because the ⁓ the analogy was we we have an email policy we have a data sharing policy Right. We we told our sales and our you know accounting folks and we told our development folks what you would not share in an email to an external party, you don't put into Chat GPT at the time. Right. I mean it it was it was just a really simple like it is the same. You're not gonna send it using you're not gonna send a corporate ⁓ financial sheet using your personal Gmail. You're not gonna put it in this system. ⁓ you're not gonna wal walk up to somebody on the street and hand them ⁓ you know, a financial document.

26So but but my point there is that we're just trying to make some common sense out of it. Right. I and I get the common sense. It's like trying to find the middle ground, you know. ⁓ I'm struggling for, you know, a real concrete example outside of the Excel example you made, because Fortune five hundred companies rely on macro heavy or just laden with formulas that half the analysts that take them over don't understand. And they are completely reliant on telling investors on a call. Like, this isn't like you have a board, you're a $30, $40 million business that's trying to grow, you've got a little private equity money, and that type of relationship. This is institutional. We're going to put this on a website. Reddit is going to light up with a bunch of stonk traders. Like hanging on every word, right? Like that's what that's an Excel models with macros built by humans, probably has a error in it, are running these companies. Now I know that I mean in an FPNA systems, this is not an unpopular opinion, so there's nothing controversial about this really here.

27FPNA systems never took off because they're not flexible enough. They're not like Excel, which is Basically just a programming interface. Yeah. Which again, you know, the idea being that's not necessarily letter of the law compliant. It's not. And so, you know, I think a lot of folks have like, you know, we we've turned a blind eye. We live in a gray world. And and and sometimes it's hard to pick and choose which parts we want to acknowledge are gray because maybe more of it than most people are willing to admit are really, really gray. And we're not tell again, right? This is not a story to say, you know, forget, you know, like anarchy and just, you know, the rules don't matter. ⁓ It's it's using some common sense and some acknowledgement of the realities of the things that we have already accepted as ⁓ okay, right? We've already accepted that spreadsheets are okay. Yeah, I mean we we had to because the work sped up so much. Like there wasn't you couldn't write that quickly. Yeah.

28⁓ you know, y y you mentioned how like, you know, we we've had people ⁓ reply in and and and say things like, you know, like, well, you know, y if you're just doing it, you know, you're not ⁓ you you don't know what it's push you know, what the AI is creating, you don't know the security and and ⁓ well, some we we do a little bit. But also, you know, the argument that I think that a lot of people are pushing forward is therefore let the professionals do their job. And trust in professionals doing their work. And ⁓ and I saw the best response to this the other day was somebody had mentioned, you know, I have never had an account or something hacked because of my poor password choices and my poor, you know, programming choices at home. but I get a letter from Citibank or Chase. or Bank of America. And so so, you know, let's just take this with a grain of salt that if you follow all the rules and you work for great big corporate Fortune fifty, that's not a guarantee of of ⁓ success and safety.

29And in fact, you know, that ten million row ⁓ you know that that got leaked out of Bank of America or one of the other banks, you know, just name ⁓ right? They've all had it happen, is probably a worse problem than you may have tomorrow with the five store franchise.