Nobody in the room is asking if the business is any good

You work out whether the family business is worth taking over the way you would assess any asset you have not yet bought: contribution margin by product and by customer, how concentrated the revenue is, whether the last five years of growth came from volume or from price, and how much of the operating result walks out of the door with the current owner. Nobody in the room is being paid to ask that.

Look at who is present when a family succession is discussed. The lawyer is there to structure the transfer. The tax adviser is there to minimise the tax on it. The bank is there to finance it. If the family is unusually thorough there is a family-business consultant, whose brief is to improve the conversation about it. Every one of them is competent. Every one of them was engaged by the person leaving.

Not one of them has been asked whether the thing being transferred is any good.

What owners say they need help with, and what the room is staffed for

The Center for Family Business at the University of St. Gallen, with Credit Suisse, studied 153 completed Swiss successions in 2022 and asked owners where they needed outside support.

Stated need for external support in a succession. Credit Suisse & CFB-HSG, September 2022, n=153.

Area of support	Share of owners
Tax advice	72%
Strategic preparation of the business	60%
Financial planning	59%
Support for the successor through the process	54%
Search for a successor	27%
Personal and emotional support	18%
No external support needed	7%

Tax leads at 72%, which is unsurprising: the tax bill is dated, quantified and unavoidable. But strategic preparation of the business sits second at 60%, ahead of financial planning, and only 7% think they can do the whole thing alone. Owners know the company itself needs work before it changes hands. That is what they say they need, not evidence of what they buy, and I would rather say so than let the two be confused. What I usually find is that the gap between them is where the bad handovers are made.

Gross revenue is the least informative number in the file

A successor is normally shown turnover, an operating profit line and a growth chart. All three are averages of things that behave completely differently.

Revenue is a blend. Inside it are products that carry the business and products that are carried, customers who pay properly for what they receive and customers quietly subsidised by the others. What settles the question is contribution margin — price less the costs that actually vary with the sale — line by line for products and for the largest customers. Not gross margin as reported. Contribution margin rebuilt honestly, including the delivery, the service hours, the returns and the discounts agreed verbally that never reached the price list.

In a founder-led company the shape is usually the same: a small group of products and customers generating more than the whole company's profit, and a long tail consuming part of it. That is not a criticism of anyone. It is what happens when a business grows by saying yes for thirty years. But it changes the question from "is this business profitable?" to "which part of it is, and is that the part I am inheriting?"

Are the top accounts held by the company, or by a person?

Customer concentration is the first thing an external buyer tests and close to the last thing a family successor is told. The share of revenue in the top five accounts is only half of it. The other half is the basis on which those accounts stay.

There is a real difference between a customer on a three-year contract with a notice period, a customer who reorders out of habit through a procurement system, and a customer whose relationship consists of a friendship with your father that began in 1994. The first is an asset. The second is an asset with a shorter half-life. The third is not an asset at all. It is a courtesy, and courtesies are re-examined when the person they were extended to retires.

Distribution behaves the same way. An agreement with terms, territories and termination clauses transfers on paper. An arrangement that was never written down, whose terms live in the outgoing owner's memory and get renegotiated over lunch once a year, does not. It has to be rebuilt by the successor, from a weaker position, against a counterparty who knows exactly how much less they know.

Was the growth volume, or was it price?

Five years of revenue growth is not one fact. It is at least two, and they carry opposite implications. If volume grew, the company found more customers or sold more to the ones it had; that is demand, and demand repeats. If price grew, the company either improved what it sells or passed through cost inflation, and pass-through runs out. A business whose top line rose steadily through a period of input-cost inflation on flat units has not been growing. It has been indexing.

The decomposition is not technically hard: units by product line by year, average realised price by product line by year, the two multiplied back to reconcile with reported revenue. What is hard is that most owner-led companies have never done it, so the answer stays genuinely unknown until somebody spends a week inside the data.

How much of the operating result is simply the owner being there?

This is the question that sets the price, and it has better evidence behind it than almost anything else said in a succession meeting. Bennedsen, Nielsen, Pérez-González and Wolfenzon studied CEO transitions in Danish administrative data and published in the Quarterly Journal of Economics in 2007. Where a family member succeeded the departing chief executive, operating profitability on assets fell by at least four percentage points around the transition.

4 percentage points the fall in operating profitability on assets around CEO transitions where a family member succeeds the departing CEO, in Danish administrative data covering an entire economy.

The obvious objection is the right one, and the authors had it first. Families are not assigned their succession route at random. A family might keep the business in the family precisely because it is easy to run, or hand it to a professional precisely because it is in difficulty. If that drives the pattern, comparing family-run and outsider-run firms tells you about the decision, not about its consequence.

So they used something that strongly predicts whether a family member takes over but has nothing to do with how well the company is run: the sex of the departing chief executive's first-born child. Firms whose first child was a boy were far more likely to appoint a family CEO, and whether a first child is a boy is close to a coin toss. It is not caused by the company's margins, its market or its management.

That gives variation in family succession which the state of the business did not create. Estimated that way, the effect comes out larger than the simple correlation — and that matters more than the headline figure. If the raw comparison had been flattered by families keeping only their easiest businesses in-house, correcting for selection would have shrunk the estimate. It grew. The correlation was understating the cost, not manufacturing it.

Four percentage points stays abstract until a balance sheet is put under it. Take an illustrative company with CHF 8,000,000 of operating assets — an example, not a finding. Four percentage points on those assets is CHF 320,000 a year; held five years, before compounding, CHF 1,600,000. That is the order of magnitude at stake in who the successor is, and it appears nowhere in the transfer paperwork.

The problem is not family ownership. It is succession without selection.

If the evidence stopped there it would be an argument against families, and it is not.

Pérez-González, writing in the American Economic Review in 2006 on US data, found that firms promoting family chief executives underperformed on operating profitability and on market-to-book. Then he looked inside the result. The underperformance is concentrated in firms whose incoming family CEO did not attend a selective undergraduate institution.

That changes what the whole literature means. The damage is not attached to being related to the founder. It is attached to having been given the job without ever having been chosen over somebody else. A selective university is a crude proxy — one filter passed at eighteen, and plenty of excellent operators never went near one. But as a marker for "this person has been assessed against a field and came through it", it separates the sample. Where the marker is present, the penalty largely is not.

The honest answer is sometimes no. And no does not mean liquidate — it means the price is wrong, or the terms are wrong, or the route is wrong.

What the successor's side of the table looks like afterwards

The same Swiss study asked successors what happened after the handover — a question rarely put to anyone while the deal is still being drafted.

• 44% said the company was strongly or moderately dependent on the predecessor. Only 32% reported low dependency.
• 30% said the predecessor could not let go: 15% fully, 15% partially.
• 49% of predecessors were still present two or more years after the handover, and 11% were still working more than 40 hours a week.
• 27% experienced open conflict during the handover.
• 42% of successors felt obliged to maintain the predecessor's approach after the transfer.

Satisfaction was bimodal, and the shape is the informative part: 51% rated the outcome 9 or 10 out of 10, and 13% rated it 1 to 4. Successions do not end up moderately fine. They end up either very good or quite bad, which is what you would expect of an outcome driven by a few structural decisions rather than by effort.

Notice what 42% describes. Not a legal restriction. An obligation felt. If the incoming owner believes they are not permitted to change the commercial model, then whatever the contract says about control, "is this worth taking over?" has to be answered on the assumption that the model stays as it is. That is most of what I have written about elsewhere as the next-generation trap.

There is one more piece of evidence worth putting next to that, with a warning attached. UBS and the St. Gallen centre published a short study in March 2026 asking potential successors themselves. The sample is 52 people — far too small to generalise from, and I present it as an indication rather than a finding, because 52 respondents in one country is a signal and not a measurement.

With that stated: only 19% said they were certain they would take over. Of the barriers named, 33% were "cannot" — capability or resources they do not have; 25% were "will not"; and 23% were "not allowed", meaning family or structural constraints. 71% said there were multiple potential successors in the family, and 63% faced a requirement that the decision be reached by consensus.

If that shape survives a larger sample, the number to watch is 23%. A candidate who is willing and able but structurally blocked is not a succession problem. It is a governance problem wearing a succession problem's clothes, and preparing the candidate harder will not move it. The version that presents as a stalled next generation is a piece of its own.

What this means in practice

A successor who has done this work arrives at one of three positions, and all three are useful. The first is yes at the stated terms: the margin is where it is claimed, the customers are contracted, the growth was demand rather than indexation, and the operating result survives the owner's departure. What remains is timing and financing, which are solvable.

The second is yes at different terms, the most common outcome and the least discussed. If a third of the operating result rests on the outgoing owner's personal relationships, then a third of it is not being sold. It is being lent, for as long as those relationships hold, and a price built on reported profit is a price for something that will not exist in three years. That is not a reason to walk away. It is a reason to move part of the consideration behind the numbers actually delivered after the handover — ordinary in any external transaction, and oddly rare inside families.

The third is no, and no is not a verdict on the family. It means this business, at this price, on these terms, is a worse use of the next fifteen years of one person's working life than the alternatives are. Sometimes the route that serves the family better is a sale, or a buyout by the management team who already run the operation — routes with their own arithmetic, above all in how long each takes, which I have set out in the piece on choosing between a handover and a sale.

The version of "no" that costs the most is the one nobody said. A successor takes on an unfit business at a fit business's price, spends fifteen years repairing something they did not break, and calls it duty.

What this cannot tell you from the outside

None of the above is an answer about a particular company, and it is worth being precise about why. The Danish result is Danish, drawn from administrative data across a whole economy, and it describes an average — averages contain firms that improved after a family handover. The Pérez-González result is from US listed companies, larger and better documented than the businesses most of this is written for. The Swiss studies are the closest fit and the smallest: 153 completed successions in one, 52 potential successors in the other. None of them knows anything about your customer list. The generational survival statistics usually quoted alongside them are worse still, and I have taken those apart in a separate piece.

What would settle it is a fortnight's work inside the company's own records. Contribution margin rebuilt by product and by customer. The top twenty accounts traced back to the contract or the relationship they actually rest on. Five years of revenue split into volume and price. An honest estimate of which parts of the operating result depend on one person continuing to answer the phone.

Here is the uncomfortable part. Every one of those facts already exists inside the business, in the order history, the invoices, the price lists and the ledger. Nobody has assembled them, because the accountant is engaged to report the past correctly, the lawyer to make the transfer valid, and the broker to complete a transaction. An honest view of whether the asset is worth having is not any of their jobs, and none of them has been asked for one.

The successor is usually the only person at that table whose entire working life turns on the answer. They are also, almost always, the only person who has commissioned no work at all.