I was reading a piece in The Economist on span of control — how many direct reports a manager should have. It’s a question, the piece notes, that has been generating confident answers for over a century. The confidence, it turns out, has always outrun the evidence.

Management thinking has always had a weakness for false precision. Clean numbers that dissolve the messiness of actual organisations into something you can defend in a meeting.

Henri Fayol, a 19th-century mining engineer turned management thinker, said fewer than six. Someone later decided seven, plus or minus two — borrowed, apparently, from entirely unrelated research into short-term memory. The number keeps shifting. The confidence never does.

I’ve watched this play out up close. A boss arrived into a new role — new to the organisation, new to the context, new to most of the people — and within weeks announced a benchmarking exercise. We were going to measure our span of control against industry peers.

Benchmark against what context, exactly? Our work was modular in places and deeply interdependent in others. Our strongest people ran themselves. The culture was built around autonomy.

The conversation ended the way those conversations tend to end.

We did the exercise. A number came back. A deck was made. A recommendation was presented. The org shifted toward a shape that made sense on a slide and considerably less sense in practice.

There’s a name for this

DiMaggio and Powell called it mimetic isomorphism — the tendency of organisations, when facing uncertainty, to copy what others are doing. Not because the evidence says it works. Because sameness feels safe.

A number gets borrowed, dressed in the language of best practice, and presented as analysis. Nobody has to defend the reasoning, because the reasoning is: everyone else is doing it. It is not a performance decision. It is a legitimacy move. The benchmark is not there to find the right answer. It is there to make the decision defensible.

I didn’t know any of this at the time. I just knew we were solving for the wrong thing. My boss was new, the pressure to demonstrate early competence was real, and a benchmarking exercise is a visible, structured, credible-looking thing to do.

I understand that now in a way I didn’t then. The impulse wasn’t laziness or bad faith. It was a very human response to an uncomfortable situation. Mimicry as a coping mechanism. The research just gives it a name.

The variables that actually determine a good span of control — the nature of the work, the capability of the people, the texture of the culture — aren’t complications to be set aside. They are the answer. There is no number underneath them waiting to be uncovered.

A number without that context isn’t a standard. It’s a placeholder for thinking that hasn’t happened yet.

_____________________________________________________________

If this resonated, you might find The Checklist Trap worth a read — on how management thinking packages complexity into something that fits on a slide.