The governance almost nobody builds

Thirty-five percent of family offices have a defined succession plan for the family office itself. These are institutions built specifically to carry money and businesses from one generation to the next, and two out of three have not worked out what happens to them when the person who set them up stops.

The survey, and the awkward part of it

UBS surveyed 307 family offices across more than thirty markets between January and March 2026, covering families with an average net worth of 2.7 billion dollars and an average of 1.3 billion under management per office. Not small operations, and not amateurs.

The investment side reads as you would expect from that population. Sixty-eight percent have formal processes for measuring financial performance. Sixty percent run an investment committee. Those are respectable numbers and they describe organisations that take the money seriously.

The governance side does not match. Fewer than half have a formal governance framework with board-level oversight. Only 35 percent have a defined succession plan for the office itself. And only 27 percent have a structured process for preparing the next generation for the roles they will eventually hold.

Set against each other the shape is hard to argue with. The top two are about the money. The bottom three are about the people who will one day decide what happens to it.

Every bar in that chart was drawn by the same people, in the same organisations, with the same resources. The difference between the top and the bottom is not capability, it is that the top half has a deadline and the bottom half does not.

There is an obvious explanation and it is not incompetence. Measuring returns is urgent, legible and satisfying, and somebody is accountable for it every quarter. Deciding who inherits authority is none of those things, it involves conversations nobody enjoys, and no quarter ever arrives in which it becomes the most pressing item. So it does not get done, in two thirds of cases, by people who are demonstrably good at everything else.

What governance actually means here

The word is unhelpful because it sounds like paperwork. In practice it answers four questions, and a family that cannot answer them in a sentence each does not have governance regardless of what is in the binder.

1. Who decides, and up to what size

   What can the office do without asking, what needs a family sign-off, and what needs everybody. Written as thresholds and categories, not as a general principle about consultation. The absence of this is what turns a single disputed deal into a two-year argument.

2. How disagreements get resolved

   Not whether they will happen, because they will. Who breaks a tie, whether that is a person or a vote, and what happens when a family member fundamentally objects to a decision that has already been made.

3. How somebody joins, and how they leave

   The route by which a next-generation family member becomes involved, and the mechanism by which anyone can exit with their share without forcing a sale of things nobody wants to sell. Liquidity for one person is the pressure that breaks otherwise sound structures.

4. What the money is for

   The most skipped and the most load-bearing. Is this capital being preserved, grown, spent, given away, or used to own and run businesses. Every allocation argument a family will ever have is a proxy for this question, and answering it directly removes most of them.

Notice that none of these are investment questions and all of them determine investment outcomes. A family that has not agreed what the money is for cannot agree on a hurdle rate, cannot agree on how much illiquidity is acceptable, and will relitigate both on every deal.

Most family disputes presented as disagreements about an investment are disagreements about the purpose of the money, arriving in disguise.

Why this matters more when you own companies

Governance gaps are survivable while everything is in liquid assets. A portfolio can be split, a position can be sold, and a family member who wants out can be paid without anybody else's life changing. That flexibility is what has allowed a lot of families to get away with the 65 percent that has no succession plan.

Owning operating companies removes it. A business cannot be divided among three siblings who disagree. It cannot be partially sold at short notice to fund somebody's exit. It has employees whose jobs depend on decisions getting made, and customers who notice when they do not. The move toward direct ownership across the sector, which is well underway, is therefore also a move toward a structure where the governance gap has real consequences.

It is worth following the mechanism through, because it is specific and it is the one that actually destroys these structures. One family member needs money, for a divorce, a house, a business of their own, or simply because they no longer want to be in this. They are entitled to their share. The share is a percentage of a company that cannot write them a cheque.

What follows has only four possible endings, and three of them are bad:

• The family buys them out with cash it has, which works and is the reason to hold some
• The company borrows to fund the buyout, so an operating business now carries debt raised for a family reason, and every subsequent bad year is sharper
• The company is sold, in whole or in part, on somebody else's timetable, which is the exact thing permanent ownership was supposed to avoid
• Nothing happens, the person stays trapped in an asset they want out of, and the relationship does the breaking instead

Which ending you get is decided years earlier, by whether anybody wrote down how somebody leaves. A valuation method agreed in advance, a right of first refusal for the other family members, a payment period long enough that the business can fund it out of earnings, and a reserve held for exactly this. None of that is complicated and all of it is impossible to negotiate fairly once one person urgently needs the money, because at that point the terms are being set between people whose interests have just diverged.

This is the same fact that shows up on the other side of the table when a family buys something. Sellers of good businesses care who takes over, and a buyer who cannot explain who will be making decisions in five years is a less attractive buyer than one who can, price being equal. Governance is not only internal hygiene, it is part of what you are offering.

The minimum viable version

The reason two thirds of these do not exist is that the full version is daunting: constitutions, councils, family assemblies, facilitated retreats, and an advisory industry ready to sell all of it. That is the right answer for a family of forty across four generations. It is a good way to ensure that a family of five never starts.

The minimum that genuinely works is short and can be written in an afternoon.

• One page stating what the money is for, agreed by everyone who has a claim on it
• A decision table: what the office does alone, what needs sign-off, what needs unanimity, with numbers attached
• A named tie-breaker, and the circumstances in which they are used
• A written exit route with a valuation method agreed in advance, before anybody wants to use it
• A named successor for the person currently making the decisions, and a date by which they will have shadowed the role
• An annual meeting with a fixed agenda, minuted, that happens whether or not there is anything urgent

The written exit route is the clause people avoid and the one that prevents the most damage. Agreeing a valuation method while everybody is on good terms costs an afternoon. Agreeing one while a family member is trying to leave costs a great deal more than that, and frequently costs the asset, because the only way to establish a price everyone accepts turns out to be selling it.

The next generation problem

The 27 percent figure deserves separate attention, because it is the one with the longest fuse. Only around a quarter of these families have a structured way of preparing their heirs for the roles they will take, and 29 percent name insufficient financial or governance education as a challenge they are already facing.

The failure mode is specific and it is not about spoiling anybody. It is that the founding generation, who built the operating company and understand exactly how the money behaves, hand over to a generation who have only ever experienced the money as an outcome. They have seen the distributions and the statements. They have not seen a bad quarter, a covenant conversation, or what it feels like to decide whether to make people redundant.

The practical version is unglamorous: give a next-generation family member responsibility for something real and small, with a budget and an outcome they own, and let them get it wrong while the stakes are survivable. That is a considerably better preparation than any amount of reading, and it is the one thing an outside firm cannot be hired to provide.

What I would do first

If a family is starting from nothing, the order matters and the first item is not the document.

Start with the purpose conversation, because everything else is downstream of it and because it is the only part that cannot be delegated. Then write the decision table, since it is concrete and produces immediate clarity. Then the exit route, while nobody needs it. The constitution, the council and the rest can come later or never, depending on how many people are involved.

What matters is that something exists, is written, and is revisited. A one-page agreement that everybody has read beats a forty-page constitution that lives in a drawer, and the survey suggests the realistic comparison is not between those two but between the one-pager and nothing at all. How the running costs and structures fit around this is in what a family office actually costs, and the discipline that governance is ultimately there to protect is the allocation decision itself. If a family is weighing its first operating purchase, the screen I would run is the same one a committee should demand.

All governance and succession figures from the UBS Global Family Office Report 2026, surveying 307 family offices across more than 30 markets between 22 January and 30 March 2026, representing families with an average net worth of USD 2.7 billion.