My nine-year-old son was recruited this season to the local football league club's centre of excellence. His coaches exude knowledge, enthusiasm and a remarkable commitment to a part-time, barely rewarded, role. They spend much of their spare time watching local junior league matches. They also handle representations from ambitious parents, other coaches and players themselves.

As a result they have in their heads a database of boys they want, don't want but will keep an eye on, and definitely don't want. Even more remarkably, they remember these boys by name. What an example of succession planning!

And then there's us. At the top of trusts like King's College Hospital foundation trust, the use of headhunters to replace executives, and increasingly non-executives, seems almost a conditioned response. So each recruitment exercise is a unique event starting from a position of ignorance.

Why? One certainty is that executives will leave. Another is that the good ones tend to know each other.

Should we not task our human resources functions with the scouting and database activities of my son's coaches? It would certainly be cheaper. Come to think of it wasn't the Department of Health having a go at something like this a few years ago? My guess is we hope the succession issue won't happen too often. Certainly, I'm guilty of not taking desultory conversations on succession forward into process.

I get away with it because of the commitment of my fellow executives to the trust. In preparing a talk recently about how King's works, I estimated that the executive directors so far total 96 years of service. Not all my colleagues welcomed this description of gerontocracy. But since it combines with a usually open mind it has mostly worked. There is an ethos - which we sometimes refer to half-jokingly as the King's Paranoia - into which the occasional new recruit can be introduced.

Succession is about blending with the team as much as finding the best candidate. The parallel with my son's coaches comes to mind again. We told one of them about someone we parents thought was a highly recruitable player. He watched, deliberated, and told us he was unsuitable because while skilful he did not know what to do with the rest of the team.

Interesting when I reflect on times when I have witnessed or heard about the brightest of executives becoming dysfunctional and falling out with each other. It is not only prima donna consultants who circle each other like strutting peacocks.

When an executive change looks likely at King's we will need a list of potential successors who can play, and who can play in a team.

What happens if we go down a level? Our general managers represent what Aston University and the Institute of Work Psychology categorised as middle managers. In the course of a study they identified female middle managers as among the most stressed employees of a sample of trusts.

Not surprising. Middle managers are the point at which performance targets and professional and shop-floor ambition collide. But how many healthcare organisations operate or commission programmes for middle management development and succession? It's something we have only got round to in a structured way in the last two years. Given the changing policy environment, it has been an absolute necessity. It raises interesting questions for succession.

As we move to trading accounts (exploiting our activity-based costing system) and as that is linked to business planning, strategy and our service change programme, we are building a mix of tools and competencies which are important to us and highly valuable to others. No surprise therefore that we have had a recent flurry of interest in our general managers from a number of colleague trusts. This gives us a succession equivalent of a double whammy.

Many of our general managers have, like the executives, been with us a long time. I appointed two of them during my first spell at King's from 1992-95. So we are unprepared for departures.

Second, there will not be an abundance of external replacements with access to the same tools and competencies. Our succession planning must focus in part on the next tier down, our business and nurse managers. But then the same dilemma will be true for them.

So besides the King's ethos we will be inducting newcomers into our business methods. Service-line reporting will close that gap, although our Salient activity-based costing system goes down to patient level so it will still be a new experience.

You could argue we are on safer ground with clinical directors, given the tendency of consultant practitioners to stay in the trust of first appointment (although that is no longer inevitable).

There is great enthusiasm here for understanding trading accounts, and time is willingly given to the potentially tedious task of identifying what costs go into what buckets. There is similar enthusiasm for improved understanding of the other technical intricacies of performance management.

But this has to reconcile with factors influencing succession. Potentially, rational management meets buggins's turn, clinical credibility, 'won't upset anybody' or other factors influencing who comes forward. When only one candidate emerges it is pretty clear that a form of selection has already taken place.

Trouble is, how to work in teams, how to influence in the subtle ways needed in multi-professional organisations, how to relate to the many other elements of a complex system are discussions after the fact.

We have to identify potential candidates and get them preparing for the role. chief executives, HR directors and medical directors as football scouts? Probably not in any current job description.

Malcolm Lowe-Lauri is chief executive of King's College Hospital foundation trust.