Structuring travel for a family office. A practical guide for the person who holds the diary.

If you run operations for a single family office, or you are the chief of staff or executive assistant closest to the principal, travel is probably not in your job title and is certainly in your week. It has an unusual shape: high value, low volume, unforgiving of error and almost impossible to delegate, because the knowledge that makes it work sits in your head rather than in a system.

The aim of a good arrangement is not to save money, although a tidy one usually does. It is to reduce the number of people who need to be told anything, to make the exceptions cheap to handle, and to survive your own holiday. This guide sets out the four decisions that do most of that work. We arrange travel for family offices, so we are not a neutral party; we have tried to write it so that it is useful whoever you end up appointing.

The single biggest cost in family office travel is not a line on an invoice. It is the repeated briefing: the gate code, the dog, the preferred entrance at the London house, the fact that the principal does not talk in the morning, the school's rule about who may collect. Every new supplier and every new driver resets that to zero.

So the first decision is how wide a perimeter one contact should cover. There are three honest answers: ground transport only, with aviation and marine handled by specialists you already trust; ground and air, so that a slot change moves the car as well; or everything, including yacht arrangements and the cars at the far end, on one itinerary and one invoice.

The right answer usually depends on how much the family moves rather than on how much it spends. A family in one city with four flights a year is well served by good suppliers kept separate. A family with homes in three countries, a boat in the summer and children at school in a fourth is not: every join between suppliers is a place where somebody has to be told something twice, and that is where the failures happen.

Travel data is unusually revealing. A year of journeys tells a reader where a family lives, where the children go to school, which building the lawyers are in, when the house is empty and, if the addresses are specific enough, a great deal that no family office would knowingly hand over. It is worth treating with the same seriousness as the bank mandate.

Four things are worth settling at the outset, and all four are reasonable to ask any supplier for. What is recorded about a journey and for how long it is kept. Who inside the supplier can see it, and whether the drivers assigned to the family are a defined group or whoever is free. Whether journeys can be booked and invoiced under a reference rather than a name, so that a leaked invoice is not a leaked itinerary. And whether the company and the individual drivers will sign a confidentiality agreement without a negotiation.

Ask also what happens to the information when the arrangement ends, which is the question people forget. A supplier who cannot answer these clearly is telling you something, whatever the rest of the proposal looks like. See security and discretion for how we answer them.

The first is the difference between principals and everybody else. A household of any size generates a great deal of travel that is not the principal: staff to and from the house, a housekeeper to an appointment, a guest to a hotel, a relative to an airport. Trying to give all of it the same treatment is expensive and slightly absurd. Refusing to arrange any of it is worse, because then it is arranged badly by somebody with no time. The workable answer is usually one account with an agreed and written difference in service, so that nobody has to make a judgement on the phone at seven in the morning.

The second is children. School runs and holiday movements are the most sensitive journeys a family office arranges and the ones most often handled by whoever happens to be available. The questions worth asking any supplier are narrow: will the same small group of drivers do it, what is the procedure if the school will not release a child to a driver it does not recognise, and who is told if a pick-up is late. A supplier who treats those as awkward questions is not the right supplier for a school run.

Long haul is easier, because it is planned. The one thing worth building in is slack on the return: a family that lands at six in the morning after eleven hours in the air does not want to discover the car was booked against the scheduled time rather than the actual one. Ask how flights are tracked and how long a car waits before anybody asks you a question about it.