When you go to buy a business, the conversation is always about the owner. He’s of age. He’s tired. He wants to take some chips off the table and go fishing. That’s the headline risk, and every buyer underwrites it.
But the owner isn’t the only one with a foot out the door.
Your head of sales has been “thinking about retiring” for four years and is one bad physical away from making it official. The plant manager who actually knows how the place runs is the same vintage. And when you look at the bench, the people who are supposed to step up, there’s nobody home.
Here’s what makes this more than the demographic hand-wringing in every fund deck. The owner’s exit is the one piece of succession risk the market actually prices, because it’s legible: it’s in the CIM, the banker plans around it, you can diligence it. The layer underneath him is invisible. It doesn’t show up in a teaser, it shows up on the floor, in week three, when you realize the only person who knows the top ten accounts is also the person closest to retiring. The market underwrites the man it can see and misses the two it can’t. That’s not just a risk. It’s a mispricing.
You see it the moment you walk the floor. A handful of guys in their twenties. A wall of guys in their fifties and sixties. In between, where the next set of leaders should be, almost nothing. This is sharpest in exactly the industries that can least afford it: chemicals, fertilizer, distribution, trucking, industrial sales. The unsexy, essential corners of the economy that everything else quietly runs on.
I want to be clear about which layer I mean. The floor has a succession plan that includes lots of robots and automation. But someone has to decide to buy them, finance them, install them, and run them. And long before any of that, someone has to sell the plant’s capacity in the first place. You don’t automate a line that has no orders on it. The machines don’t fill their own backlog.
Here’s the irony, and it’s the same root cause cutting both ways. The professions that built real succession machines, like banking, law, and consulting, did it because the work was legible enough to systematize. You can write down what a first-year associate does, so you can train one, promote one, and replace one. But legible work is exactly what the technology eats first. The physical economy’s management layer never got written down. Reading a customer, knowing which competitor is about to stumble, closing on a handshake, deciding where the next plant goes and what it should make, none of it ever fit in a training manual. So it has no bench. And for the same reason, it’s the work I don’t think AI touches in any real way for years. The thing that left these companies undermanaged is the thing that makes them hard to automate.
So when you buy that business, don’t just underwrite the owner’s exit. Look one level down. Then look again. Price the bench the way you’d price deferred maintenance on the plant, because that’s what it is: a liability nobody put on the schedule. The threat is bigger than the one you came in to size up.
You’re not buying a succession plan. In most cases, you’re buying the job of building one, for the whole company.
the counterfactual 