Succession is usually treated as an appointment. A date is set, a candidate is chosen, a transition period is agreed, and the organisation braces for a handover measured in months. The appointment is the visible part. It is not the part that determines the outcome.

What actually transfers

A leadership role is held together by things that do not appear in a handover note: which relationships are load-bearing, which commitments were made verbally, which numbers the board trusts and which it quietly discounts. None of that transfers in a document, and very little of it transfers in a six-month overlap.

It transfers by exposure. The successor has to have been in the room — not observing, but owning something that could go wrong — for long enough that the relationships are theirs rather than inherited.

Three years, not three months

The useful window opens about three years out. That is enough time for a successor to run a full cycle of something material, to be seen to make a decision that was unpopular, and to build a direct relationship with the board, the largest customers and the regulator that does not route through the incumbent.

It is also enough time to discover that the intended successor is the wrong choice while there is still room to do something about it. That is not a failure of the process. It is the process working.

The incumbent is the constraint

The hardest part of succession planning is rarely the candidate. It is the incumbent, who is being asked to give away the parts of the role that make it worth holding, several years before leaving, with no guarantee of what comes next for them.

Boards that handle this well address it directly: they agree what the outgoing leader keeps, what they hand over and when, and what their position will be afterwards. Boards that avoid the conversation end up with a successor who has the title and an incumbent who still has the relationships — which is the most common way a well-planned succession comes apart.