The most common type of family office has no staff, no premises, and nobody who would call it that.

It is a founder, a long-serving assistant, an accountant, a lawyer used for twenty years, and an investment adviser spoken to each quarter. All of it coordinated by one person, usually from a phone and an inbox, in the gaps between running everything else.

Plenty of people in that position tell me they do not have a family office yet. In reality, they have had one for years. It just is not the version in the consultancy brochures.

I keep having that conversation, and the reason is that we use one phrase for at least six different things. Knowing the difference helps you see the mechanics, and work out which model is yours now, and which one you are heading towards.

The six I see in practice:

→ The invisible one. The founder is the office. A good assistant, a trusted adviser or two, everything routed through one person's head. Maximum control, almost no overhead, and a single point of failure with the founder's name on it.

→ The embedded one. It lives inside the operating business. The company's finance and admin team pays the household bills, keeps the records and handles the founder's needs. Fast, already staffed, and sensible right up until the business and the family start pulling in different directions.

→ The administrative one. Built to run the life, not the portfolio. Several homes, staff, heavy travel, maybe a boat or a plane. The logistics are the job.

→ The investment-led one. The version everybody pictures first. It behaves like a fund, with an investment team and the infrastructure to match. Usually built by a founder who came from investing, or by assets large enough that deploying them needs an institution.

→ The multi-generational one. Built to outlive the person who made the wealth. Several branches, several generations, and a real output that is continuity rather than return. A dynasty, in the proper sense of the word.

→ The philanthropic one. The mission is the mandate, and the work is deploying capital into impact.

Six models. Different costs, different hires, different systems underneath. Compare them to each other under one label and you will reach the wrong conclusion, which is how someone decides they have nothing when they have been running one for nine years.

Here is what I find more interesting than the list.

Almost no family office starts with an architecture and a launch plan. They get built one decision at a time, and each decision is a good one when it is made.

The investment-led office exists because the founder came from investing. The embedded office exists because the finance team was already sitting there and someone had to pay the bills. The administrative office exists because the third house and the travel made it unavoidable. The invisible office exists because the founder has always been the person who knew everything, and nobody ever decided to change that.

That is how most good things get built. Then the wealth grows, the complexity grows with it, and all those sensible individual decisions start to look different in aggregate.

It is like building a bungalow, and then deciding to put four floors on top of it. Nothing about the bungalow was wrong. The foundation was simply poured for a different building.

You can usually see where the strain will show.

The embedded office breaks at the liquidity event. The team that ran everything belonged to the business, and the business has just been sold.

The invisible office breaks when the founder runs out of capacity, or health, or patience. It works perfectly until the person in the middle is unavailable, and then it does not work at all.

The investment-led office tends to strain at the second generation, when people who did not build the wealth inherit an institution designed around somebody else's appetite for risk.

Every one of those is foreseeable. Almost every one arrives as a surprise.

The one thing I have come to understand is that underneath all six, every version is trying to do the same job. Hold complexity together and coordinate across domains that do not naturally talk to each other. Tax, legal, investment, property, people, giving.

The shape changes. The coordination does not.

Which is the part worth holding on to if you are at the start of this. A family office is not a building or a headcount. It is a discipline, and it can be practised in many different styles depending on what the family needs.

So when families ask me how to organise their wealth, I do not start with the model. I start with where the complexity actually sits, and where they are spending time they do not want to spend. That alone tells you half the story.

If you want to know which office you are really running, do not look at job titles. Pay attention for a month to what reaches you and what it is about. If most of it is property, staff and travel, you are running an administrative office whatever you call it. If it is managers and allocations, you are investment-led. If it is branches and distributions, you are multi-generational and probably underbuilt for it.

And if the honest answer is that all of it reaches you personally, you are running the invisible version. That is the most common of the six, and the only one that stops working the moment one person is unavailable.

Then ask the second question, which matters more if you are building something now. Where will the complexity sit in five years. A sale, a move, a new generation coming in, a shift from building wealth to passing it on. Each of those changes which model you need.

You will not get the design right first time, and you do not need to. What you need is a structure you can extend without knocking the whole thing down. Pour the foundation for the building you are likely to end up with, not just the one you need this year.

P.S. On 13 October at 12:00 PM, we are running an expert session on why so much succession planning stalls before it gets anywhere. I have been looking forward to this one all summer.

We're joined by Catherine Grum, who has spent more than twenty years in this world at Barclays, KPMG and BDO, and who now advises families independently on family office design, governance and succession. She has seen more of these conversations go well and badly than almost anyone I know.

It is an intimate, invite-only session, so seats are limited. You can reserve one here.

Until next week.

— Amin Naj