2 Years to $40m: The Anatomy of a High-Velocity Roll-up

If the calls are simple, like I want my invoice, the AI would take the invoice from the system, confirm with the client all the details before sending it, and then it sends it. If the request is complex, like a problem with the fire safety installation, it passes directly to the person that can solve that. In this case, we were AI reactive because we built a call center actually last year.

But now we look at all the problems before changing org, before outsourcing, before changing the configuration of the teams, if we can automate it, right? But this is another example with AI. It's actually being super good for cost saving, but also, and more importantly, for our customer experience, right?

So talk to us about the level to which you're integrating and what challenges, what people might learn from that.

Yeah, so as you say, AI and in general, you know, automation is a way of turning a services business, which most of the costs are labor, into a high operating leverage business where you do CapEx and then you replace manual labor by something that scales better, right?

So for us, the whole thesis revolves around, as I mentioned, on the revenue per technician, the ability to really maximize the amount of work that a technician can do safely on the hours that they work. And so for us...

uh density of routes and customers per region it's paramount so we go into a region then we start acquiring more customer bases on that area and so we need to integrate right so for us it was never a matter of whether we integrate or not and the most important part is to integrate the customer base the route management system and the and the routing of the technicians

after that then you start to more traditional integration like back office instead of having again like 20 uh invoicing people all throughout we just put in a in an excellent center we create workflow systems but that's more traditional private equity right but that's how we decide what to integrate and what not

You actually make it sound easy, but you have told us that integration has been probably the key challenge here, because certainly finding targets and acquiring them has not been a challenge. I'm not saying it's easy, but you've done so many in such a short amount of time. It's clear that's not the bottleneck.

The integration is. So what has been such a challenge? Because what I just heard you say sounded very systematized, very dialed in.

No, I would say that, by the way, even M&A is hard, right? Like something that we have talked about, AI and systems and processes, but really it starts, it sounds cliche, but it starts from the team. So we were three partners, right?

We had a team of two M&A folks since the beginning. We have now, after so many years, five people that come from BCG, Bain, McKinsey. So we have really good team and everybody's fully heads down on their own thing.