A holding-company operator put a question to me recently that I have not been able to put down.
When you diligence a business, you study the CEO. You read the management presentation. You back-channel the references. You form a view on whether this is the leader to carry the thesis. Then you ask a different question. If that CEO left tomorrow, who runs the business on Wednesday morning?
Most deals never test for the answer. They test the person at the top and stop there. The bench underneath gets a paragraph in the management section and a polite nod in the IC memo. And then eighteen months into the hold, the CEO leaves for a bigger seat, and the firm discovers there was only ever one person who understood how the company actually worked.
That is key person risk. It does not show up in the model. It shows up in year two, when the lever you were counting on walks out the door.
The question diligence skips
The standard management workstream is built around the CEO and the CFO. That makes sense. They carry the story and they carry the numbers. But the workstream tends to stop at the names on the org chart instead of testing what happens when one of them is gone.
Here is the test. Pick the CEO. Now remove them. Who steps up, on what timeline, with what authority, and how much of the operating knowledge leaves with the departure?
If the honest answer is “we would run a search,” you have a thin bench. The business has been running on one person’s judgment, one person’s relationships, and one person’s memory of how the last three problems got solved. That is a fragile asset to underwrite a five-year hold on.
A strong number two changes the shape of the risk. It means the company has more than one person who can hold the whole picture. It means the CEO has been building an organization rather than performing one. It means the thesis does not live or die on a single employment contract.
Why a thin bench travels with a data problem
This is the part the operator’s question pushed me toward, and it is where my work lives.
A company that depends on one person’s head tends to keep its operating knowledge in that head. Not in documented, traceable systems. In the founder who remembers which customers are really profitable. In the ops lead who knows which numbers in the report are trustworthy and which ones everyone quietly works around. In the one analyst who can actually produce margin by product line, because only they know where the real data is and how to stitch it together.
When knowledge lives in people instead of systems, two risks are present at the same time. Lose the person and you lose the capability. And you could not see the capability clearly in the first place, because it was never written down in a form anyone else could read.
A strong number two and a legible data layer are two signs of the same thing. Institutional maturity. They both answer the same underlying question. Does this company’s ability to operate exist outside of any single individual?
A mature company has distributed its operating knowledge. The bench is real because more than one person carries the picture. The data is legible because the business decided, at some point, that the truth should live in systems rather than in someone’s recall. The two travel together because they come from the same instinct. Build the institution, not the dependency.
A fragile company has concentrated everything. One leader, one source of truth, and that source of truth is a person. When you find a thin management bench in diligence, look at the data layer. You will usually find it is thin in the same way and for the same reason.
This is also why hiring your way out of the problem rarely works on its own. A board that spots the gap late often reaches for a senior hire to plug it, the same move I see firms make with data leadership when they hire a CDO before they have made the decision. The new person inherits the concentration rather than dissolving it, because the knowledge they need is still locked in someone else’s head and the systems still cannot produce a number without that someone. You do not fix a single point of failure by adding a second person on top of it. You fix it by distributing what the first person knows.
What buyers should test
If you are on the buy side, the management workstream needs a few questions it probably does not ask today.
Run the Wednesday test out loud with the CEO. Ask them directly. If you were hit by a bus tonight, who runs this company tomorrow, and what would break first? A confident CEO with a real bench answers this in specifics. A CEO who has not built one gets uncomfortable, because the honest answer is that a lot would break and they know it.
Ask the number two to walk you through the business without the CEO in the room. Not the polished management pitch. The actual mechanics. How decisions get made, where the numbers come from, what the team would do if a major customer churned next quarter. You learn very quickly whether there is a second person who holds the whole picture or whether you are talking to someone who executes the CEO’s instructions.
Test how long it takes to get a number that matters. Ask for margin by product line, or revenue by cohort, and watch who gets pulled in and how long it takes. If the answer requires the one analyst who knows where everything is, you have found both the data risk and the people risk in a single pull. The capability is real, but it is sitting in one person.
Look at where the operating knowledge lives. Are processes documented or remembered? Is there a single source for the metrics the board tracks, or does each leader bring their own version? A business that has written itself down is a business that has more than one person who can run it.
The time it takes a company to answer these questions tells you a lot about how it will perform under the pressure of a hold and an exit. That is the same instinct behind the Diligence Clock. The speed and confidence of the answer is the signal, not just the answer itself.
What sellers should build
If you are running a portfolio company toward an exit, this cuts the other way. The strong number two is something you build, and you build it well before the buyer asks.
Two years out, name your successor and start widening their authority. Let them run board sessions. Let them own customer relationships that currently run through you. The goal is that by the time a buyer is in the room, there is a visible second person who clearly holds the business, and the buyer can see it without being told.
This is also a structure problem, not only a talent problem. A thin bench is often the symptom of an org built around one person’s gravity rather than around clear roles and ownership. I have written about how structure follows clarity, and the same logic applies here. You cannot build a real number two inside an organization where every decision still routes back to the founder.
Then do the harder thing. Get the operating knowledge out of people’s heads and into systems. Document the processes. Establish a single source and a single definition for the handful of metrics that matter, the ones finance, sales, and operations should all agree on. This is the work that makes a company legible to a buyer, and it is the same work that makes the company survivable when a key person leaves.
There is a relationship to understand here too. When data lives in one person’s head, that person tends to get blamed when the numbers are wrong, even though the real fault is that the company never built a system anyone else could use. I covered that pattern in why the data team gets blamed. Building the bench and building the data layer are the same act of taking the dependency off any single individual.
Two signs, one underlying truth
The strong number two is a simple criterion and most deals miss it. It is missed because the management workstream is built to evaluate the leader at the top, not to stress-test what happens when that leader is gone.
But it is one of the cleanest reads on institutional maturity you can get. A company with a real second-in-command has usually also built a real data layer, because both come from the same decision. Stop depending on one person and start building an institution.
So add the Wednesday question to your diligence. If the CEO left tomorrow, who runs the business on Wednesday, and can the answer be found in the systems rather than only in someone’s head? A confident answer on both tells you the company is more durable than the model suggests. A weak answer on either tells you the value is more concentrated, and more fragile, than the price assumes.