There is a person in your organisation everyone routes things to. You know who they are. So does everyone else, which is the problem.
They are competent, they answer, and things they touch come back finished. Over a few years that reputation converts into a workload with a particular shape: a great deal of it is urgent, most of it is somebody else's, and a surprising share of it is work that a considerably less capable person could have done. The reward for being reliable is more of everything, sorted by who asked most recently rather than by what matters.
Nobody is looking at this, because nothing measures it
Headcount planning tracks how many people are in a function. It does not track what any of them spend the week on, and it certainly does not track the informal escalation path that routes a disproportionate share of the hard-and-also-trivial to four or five individuals. That allocation is made by hundreds of small decisions by people acting reasonably — ask the person who will get it done — and nobody is responsible for the aggregate.
The aggregate is what costs money. A senior engineer answering integration questions all week is not architecting anything. A commercial lead cleaning up a proposal template is not in front of a customer. These are not failures of prioritisation by the individual; the requests are legitimate and each one takes twenty minutes. It is a failure of the organisation to notice that its scarcest capacity is being consumed by its least scarce work.
I want to be precise about what I am not saying. This is not an argument that senior people should refuse unglamorous work — an organisation where the experienced decline the tedious is a worse organisation, and everyone can name one. It is an argument that the routing is invisible, and invisible allocations do not get examined.
The version I find most costly is the one that looks like mentoring and is not. A capable person answering the same question from a different colleague every week is not developing anybody; they are functioning as an undocumented internal service. Written down once, that question stops arriving. Nobody writes it down, because writing it down takes an hour and answering it takes twenty minutes, and the twenty minutes is charged to a different account than the hour.
Where I would look first is the manager, and the finding is usually structural rather than personal. Seeing this requires knowing what your people actually did last week, which requires enough time per person to ask — and a manager with too many reports resolves that by routing work to whoever needs least supervision, which is the mechanism itself. It is the span of control stretched past what anyone can hold producing its most expensive symptom.
There is a simple diagnostic and almost nobody runs it. Ask your five strongest people to list what they spent last week on, and mark each item with whether it required them specifically. The proportion that did not is the number, and in every organisation where I have seen it done, it was higher than the executive expected and unwelcome enough that nothing happened afterwards. Which is a shame, because it is the only line in the budget you can reallocate without hiring anyone — and it is attention, the one resource nobody budgets.
