Some Problems Can't Be Solved in a Spreadsheet.

I have built financial models for people decisions, so I'm not making an argument against measurement.

Numbers are useful. They make cost visible, they make trade-offs discussable, and they turn a vague concern into something a leadership team can actually decide on. If you can't put an approximate figure on what a problem is costing you, it will keep losing to problems that have figures attached.

But there's a line, and it's worth being honest about where it sits.

A model can tell you that turnover in one team costs several times what it does elsewhere. It cannot tell you that the cause is a manager who is excellent at the work and struggling with the people. A model can quantify the capacity your leadership team spends on employee matters. It cannot tell you which of those matters was handled well and which one is going to come back.

This is why I deliberately refuse to put a dollar value on employee relations. Not because it's unmeasurable in principle, but because a generic number would be misleading in exactly the situations where the stakes are highest. Every one of those situations depends on the people involved, the history, and the context. Averaging them produces a figure that looks precise and isn't.

The failure mode I see most often isn't ignoring data. It's using data as a substitute for a decision. A leadership team that spends three weeks refining a benchmark analysis is often avoiding a fifteen-minute conversation about whether the right person is in the role.

The practical balance: use numbers to size the problem and prioritize attention. Use judgment to decide what to do. And notice when the analysis has stopped producing new information - that's usually the moment the real work begins.

Not everything that matters can be calculated. Not everything that can be calculated matters equally. Good leadership is knowing which situation you're in.