The Architecture of Decision – how family offices accumulate the systems they
decide inside, and what changes when they design them.
By Lance McPherson, Partner, Label R
Between 2021 and 2023, several of Luxembourg’s largest single-family offices allocated to the same three private equity managers, in the same vintage window. By 2023, those positions were being marked down in the same quarter. Each office made its own decision. The GP relationships were real. The track records existed. The diligence was thorough. The principals were comfortable. And the outcome was identical.
That is not a story about due diligence failing. It is a story about what happens when sophisticated, independent decisions get made inside the same environment.
A universal Pattern
Behavioural science has spent decades demonstrating something that holds wherever it has been tested: small structural changes around a decision produce large changes in the decision itself.
Paint a fly inside an airport urinal and cleaning costs drop. Play classical music in a petrol station and crime falls. Rearrange the order of food on a buffet line and people eat differently. The people stayed the same in every case. The environment made one choice feel more natural than another.
The same dynamics apply in private rooms where capital is allocated, governance is exercised, and generations are decided between. Those rooms are smaller and more sophisticated. That just makes the patterns harder to see.
Decision architecture
Every family office operates inside what could be called a decision architecture: the sequence, defaults, roles, and signals that shape how decisions actually get made. It is the structural layer beneath the analysis and the talent.
In most family offices, that architecture was never designed. It accumulated. It grew out of who joined first, how the founder preferred to be briefed, what the lawyers required for documentation, and which conversations happened over which dinners.
Personality, habit, and history did the design work.
Most family offices have made themselves wealthy inside that accumulation. The
challenge is that it continues. The environment keeps producing decisions long after
the conditions that built it have changed.
Independent rooms, identical outcome
Return to the offices at the top of this piece. They were well-resourced and well-led. They were also sitting inside a decision architecture they shared without realising they were sharing it.
A long-standing GP relationship became a signal of trust. A track record from a different rate environment became a signal of capacity. Peer commitments became signals of consensus. The principals were comfortable because the materials were designed to produce comfort. Each signal was real. Each one, on its own, was a defensible read.
Together, those signals acted as one. Each commitment became the basis for the
next. By the time the last office decided, what looked like independent conviction was
a queue. Each office made a defensible decision. The architecture made the decision once, and the room repeated it.
Sophistication does not protect against this
A long-standing GP relationship became a signal of trust. A track record from a different rate environment became a signal of capacity. Peer commitments became signals of consensus. The principals were comfortable because the materials were designed to produce comfort. Each signal was real. Each one, on its own, was a defensible read.
Together, those signals acted as one. Each commitment became the basis for the
next. By the time the last office decided, what looked like independent conviction was
a queue. Each office made a defensible decision. The architecture made the decision once, and the room repeated it.
Three terrains
Decision architecture operates on three terrains in every family office. It stays
invisible on each one until you start looking for it.
Capital. Which assets to hold, sell, or move into. This is the most visible terrain, and the one that gets most of the attention.
Governance. Who speaks first in committee. Whether risk is reviewed before performance, or after. What gets shown, in what order. The simple test: in your last three committees, who in the room had the structural right to say no to the principal, not the moral right, the structural one?
Generation. What gets preserved when leadership changes. Who has the legitimacy to retire something the founder built. The simple test: if the founder stepped back tomorrow, which assets would no longer make the cut, and who would have the standing to say so out loud?
What most family offices have in common is this. Their architecture on capital is at least examined. Their architecture on governance is largely informal. Their architecture on generation is almost always implicit. And the three are entangled.
What appears to be an investment process is often a governance structure in disguise. A governance structure that protects the founder’s voice is also a generation structure that delays the next one.
From problem to design — three changes implementable this
quarter
The interesting question is what well-designed decision architectures do differently. The difference becomes visible in small structural decisions, each of which operates at the level of the environment. Their effects compound across every decision that follows.
- Reverse the order. Most investment committees review performance before risk. The order shapes what the room notices. Reverse it, and the conversation changes.
- Collect written views before discussion. The first person to speak in a committee anchors the room. Some offices collect written views before any verbal discussion begins. The room gets to disagree on paper before anyone has to disagree out loud.
- Separate legacy and new capital into different forums. When evaluated together, legacy assets almost always win because familiarity weighs more than the criteria suggest. Some offices evaluate legacy and new opportunities in separate forums, with the same criteria. The outcomes change.
An insight changes how you think. A design changes what happens. These are designs.
A fifteen-minute test
There is also a question that costs nothing. The next time you review a position in your portfolio, before you open the book, ask yourself this: if we did not own this today, would we buy it at the current price? Write the answer down. Then look at the cost basis. See whether the answer changes. If it does, that gap is not a failure of judgment. It is your decision architecture producing the answer instead of enabling one.
What the next twenty years look like
The last two decades in family offices were won by better information, better managers, better tools. Those investments mattered. They produced the sophistication the industry now relies on.
The next twenty years are likely to be won differently. The constraint on decision quality is no longer the information. It is the system that produces decisions from the information. For most offices, that system accumulated.
Most family offices try to make better decisions. The most resilient ones design the environment in which decisions get made. That work determines what kinds of decisions the room is capable of producing.
The question is no longer only whether a family office has good information. It is whether its structure allows that information to change the decision at all.
And most of that work stays invisible until you start looking for it deliberately.
Lance McPherson is a Partner at Label R, and a former Board Director of the
Behavioural Insights Team.