TL;DR
- The AI boom will bring private equity to your door. The pitch is aggregation plus AI-driven cost reduction, and it works, for about three years.
- It works for three years because every company in your category will run the same transformation by year four, and everyone's margins will converge again.
- The durable value is not the cost saving. It is the generalizable IP that comes out of solving a problem only your industry understands.
- You already have the data, the context, the peer network and the buyer's instinct. Selling the company hands all four to someone who has none of them.
What does a PE firm want from a physiotherapy clinic? Why do they roll up many of them into one?
It works just like Thrasio. Thrasio is a roll-up of Amazon merchants. According to Vista Partners (they’re like “Suits” on Netflix, but somehow worse), all companies are like chickens. They have the same few problems. The sameness means they can all be rolled into one, slapped with 500 AI-driven workflows by operating partners, and voilà: you have an AI-first PT clinic whose contribution margin is better than those of SaaS.
Thrasio has closed down. Operating partners moved on to Ramp. Ramp was the off-ramp for operator PEs. So when I hear that General Atlantic is rolling up BPOs, my mind immediately goes to: “Which VC-funded company does a GA analyst go to, and can I buy that stock on the secondary market?” OK, enough of talking like a “parody PE bro” on Twitter. Let’s talk AI.
I get the lure of aggregating companies, slapping on AI, and reducing cost. It’s marginally exciting for three years because every company will have the same transformation initiatives in four years, and their margins will all look the same.
The margin question nobody answers cleanly
Someone who is an actual PE or more educated than me should explain this: do PEs care to stay ahead of the pack in terms of margins (and hence valuation), or will simple aggregation-led scale get them decent multiples and the AI unlock is just table stakes? I ask this because AI will lift everyone’s margins.
The line of business hiding inside your operations
This brings me to my next question / thought exercise: new business lines.
At Moative, we do a bunch of interesting things that don’t neatly fit into the definition of consulting or product startups. I would call these “new lines of business,” not for us but for the companies we work with.
A case of this would be an industrial company doing something with AI that unlocks cost savings for them, but it’s so tied to understanding their engineering or production that it’s not very apparent—unlike RCM or accounting, where cost savings are easy to theorize from the outside in. One has to spend months, if not years, to realize the potential.
Why you are the one who should own it
The big win is not when you realize the cost or time savings but rather the generalizable IP that you can take to similar companies. That’s a new line of business or startup, if you will. You, the industrial company, have the context and data. But why should a startup walk away with the residual knowledge and build IP out of it?
You have the data and know-how.
You hobnob with a dozen of your industry peers every summer in Vegas.
You know exactly where the value unlock is and how to sell that value—because you are the buyer, first and foremost.
I see this. Hence I don’t see the PE game of margin optimization nearly as exciting, nor do I find the idea of funding a vertical AI startup that takes three years for product-market fit and yet won’t have the data and context.
Instead, we like helping our clients think big. They have the secret sauce. We can cook.
As any 15-year-old would say these days, “Let’s cook!”
Where Moative fits
Moative is neither a product company nor a consulting firm, which is awkward at conferences and useful in practice. We help companies with deep industry context turn an internal AI unlock into something sellable, through joint ventures and outcome-linked arrangements rather than license fees. In healthcare revenue cycle management that means the operator keeps the know-how and the upside, and we carry the build risk. The reasoning behind the structure is in Buying vs. Becoming AI.
Frequently asked questions
Why would private equity want to roll up companies in my industry?
Because sameness is the whole thesis. Companies in a category share the same handful of problems, so a fund can aggregate them, apply one set of AI-driven workflows across all of them, and lift contribution margin without inventing anything.
Does the AI margin unlock actually create lasting value?
Not on its own. Every company in the category will run the same transformation within a few years, at which point margins converge and the advantage disappears. It is a timing play, not a moat.
What is the difference between an obvious AI saving and a valuable one?
An obvious saving is legible from outside the company, which means anyone can underwrite it and everyone will capture it. A valuable one is bound up in your specific engineering, production or workflow, takes months to find, and produces IP that generalizes to your peers.
Why not just fund or buy a vertical AI startup instead?
Because a startup typically spends years reaching product-market fit without the data or industry context to shortcut the search. You already have both, plus the distribution and the credibility with buyers who are your peers.
Who owns the IP when an industrial company builds AI with a partner?
That is the question to negotiate before anything is built, and most companies do not. Standard arrangements leave the data with you and the resulting IP with the vendor, which is the wrong way round when the context came from your side.
Should a midmarket company ever sell to private equity in an AI cycle?
Sometimes, for reasons that have nothing to do with AI, such as succession or capital for expansion. The argument here is narrower: do not sell because a fund has convinced you that AI-driven margin expansion is something only they can deliver. It is not, and the part of your business that is genuinely hard to copy is the part you would be giving away.any does a GA analyst go to, and can I buy that stock on the secondary market?” OK, enough of talking like a “parody PE bro” on Twitter. Let’s talk AI. I get the lure of aggregating companies, slapping on AI, and reducing cost. It’s marginally exciting for three years because every company will have the same transformation initiatives in four years, and their margins will all look the same. Someone who is an actual PE or more educated than me should explain this: do PEs care to stay ahead of the pack in terms of margins (and hence valuation), or will simple aggregation-led scale get them decent multiples and the AI unlock is just table stakes? I ask this because AI will lift everyone’s margins. This brings me to my next question / thought exercise: new business lines. At Moative, we do a bunch of interesting things that don’t neatly fit into the definition of consulting or product startups. I would call these “new lines of business,” not for us but for the companies we work with. A case of this would be an industrial company doing something with AI that unlocks cost savings for them, but it’s so tied to understanding their engineering or production that it’s not very apparent—unlike RCM or accounting, where cost savings are easy to theorize from the outside in. One has to spend months, if not years, to realize the potential. The big win is not when you realize the cost or time savings but rather the generalizable IP that you can take to similar companies. That’s a new line of business or startup, if you will. You, the industrial company, have the context and data. But why should a startup walk away with the residual knowledge and build IP out of it? You have the data and know-how. You hobnob with a dozen of your industry peers every summer in Vegas. You know exactly where the value unlock is and how to sell that value—because you are the buyer, first and foremost. I see this. Hence I don’t see the PE game of margin optimization nearly as exciting, nor do I find the idea of funding a vertical AI startup that takes three years for product-market fit and yet won’t have the data and context. Instead, we like helping our clients think big. They have the secret sauce. We can cook. As any 15-year-old would say these days, “Let’s cook!”
