Dave Cava on Why Operational Maturity, Not AI, Determines MSP Survival

Dave Sobel (0:01)

Dave Sobel here, reporting from the floor of ChannelCon in San Diego, sitting down with Dave Cava. Now, Dave is owner of PeopleSharp and the co-author of The Pumpkin Plan for MSPs. Before that, 12 years as COO and CFO of Proactive Technologies, a New York MSP, Inc. 5007 times sold in 2019

He sits on GTIA's Workforce Advisory Council, and his session here is AI Readiness for TSP's What Actually Matters. Dave, welcome to the show.

Dave Cava (0:30)

Thank you.

Dave Sobel (0:31)

So you published a piece in July under a really blunt headline, AI isn't coming to save your MSP. So make the argument, what are MSPs getting wrong right now?

Dave Cava (0:44)

The fact is that it's over 50 percent of MSPs that are less than 5 percent profitable. And the long and the short of it is that AI will help to bury those MSPs if they don't worry about that issue more than they worry about the technology that's here and coming. There is no excuse. No, there's no substitute for operational maturity. You have to have it. Profitability is going to drive R&D. It's going to drive the ability to be out in front on AI.

The largest MSPs that are the best funded are just making leaps and bounds in the area of AI.

It's like 27 percent of MSPs that are actually losing money. There is a quiet crisis going on in the industry where the rich are getting richer and the poor are getting poorer. And I kind of see it as part of my mission to save the saveables, or can, you know, just try and wake people up. If one out of ten people listens and says, hey, I need to become a business person and not just a technologist, that's a win.

Dave Sobel (2:07)

Say a little bit more about the saveable group, because it sounds like if you think the big guys are doing it, they're crushing it, right? And how do you view the difference between like the ones that aren't making any money and maybe the saveables and where does that overlap? Like think about it.

Dave Cava (2:19)

You don't need to be big to be profitable, right? You didn't mention, I used to co-own a company called Encore Strategic, they're still going, they're here doing a great job, just did a great session.

But one of our mottos there was profit, grow, exit, and we used to talk about how it's important that it happens in that order. That if you have an unprofitable business and blow it up, you're just going to have a big, unwieldy, unprofitable business, right? And people who found MSPs are mostly technologists, generally. They're not coming out of Wharton or Harvard Business School. They're coming out of being nerds who like technology like you and me. They don't think about things in terms of business fundamentals until they need to, for the most part. I think that's starting to change with some of the people that are starting MSPs today. But for the old crusty guys like me that have been around 15, 20, 30 years, they got into it because they love the technology. And some of them just never picked up the business and they never learned how to create a business that really has increasing value, becomes a saleable asset, and is consistently profitable.

Dave Sobel (3:39)

It resonates with me. I ran a conference literally with the theme of profitable is enough.

Sometimes like that is enough to get started on the rest. Now, I want to ask a little bit, your model is the farm system. This is really interesting. Entry, higher entry level, develop, promote up. But last October, you wrote, I'm quoting this, will automation and AI lessen the need for these kinds of resources in the future? Yes. So if AI eats the bottom run, where do those L3s come from in 2033?

Dave Cava (4:11)

That's a great question. It's one that I've been asking myself. First of all, I like that you said 2033 There is a cottage industry in and around the MSP world that is always chatting the next big thing.

And for a while, it was cloud. Now it's AI.

And the fact is, AI will change everything, but it's not going to happen in six months. If it was going to happen in six months, we would have already been a completed process. We're still just starting to scratch the surface there. So we'll be gradual. And I think 20 years ago, I was talking about the concept of concierge IT support. I saw what was coming and that it was less about, you know, being the best engineer ever and more about understanding your client's business and being able to connect them with the right products and get those products to play together nicely. And I think that that's just snowballing, in a sense, that understanding your client's businesses and work flows, it's not going to be optional anymore. And I don't really think that AI is going to eat that first layer entirely for a long, long time, if it ever does. I think those level one positions, the soft skills are going to be more important. They've always been important. They're going to be super duper important to the point where somebody that doesn't have them just doesn't belong in that job. The idea is to keep the friendly face, the human element, the helpfulness, the I'm on your side, and make what that person is doing on the back end to get the tech to work a lot more efficient and repeatable and take all the complexity out of simple level one tasks.