Idea

Every function has a legitimate indicator, and it is their simultaneous sovereignty that prevents a shared definition of success

Info

Originally written in French. Translated by AI — the meaning has been preserved, not the prose.

Main idea

Marketing watches visibility or engagement, sales watches volume, conversion or field presence, product watches usage, retention and functional quality, finance watches margin, cost and return on investment. None of these indicators is wrong, and each can be defended by solid reasoning. So the disagreement is not a disagreement about the numbers.

It appears when each indicator becomes sovereign in its own corner. The organization no longer shares a definition of success: it juxtaposes several, sometimes incompatible, and each party can demonstrate that it is pursuing a rational objective. Yet the whole produces no single line, since nothing says what comes first.

What a hierarchy adds is not a decorative ranking. The main criterion says what the organization is really trying to succeed at now; the secondary criteria say what has to be preserved so that this success is not obtained at the price of destruction elsewhere. Pursuing volume can be coherent if you accept a temporary reduction in margin; pursuing margin can be coherent if you accept giving up certain segments. Ordering the metrics therefore amounts to making explicit the way the organization agrees to judge its own action.

Why it matters

This shifts the debate about indicators: the question is not which ones are relevant — nearly all of them are — but which one wins when two contradict each other.

It also defuses a frequent and misdirected criticism, the one that blames a function for defending its own criterion: the flaw is not in the indicator, it is in the absence of ranking.

Nuances and limits

A hierarchy has a short shelf life. A main criterion left in place after a change of phase — from conquering a market to seeking profitability — steers the organization against its own situation.

And ordering does not replace judgment: two criteria can come into conflict in a particular case no written hierarchy had anticipated.

Open questions

  • At what frequency and on what trigger should a hierarchy of criteria be replayed so that it keeps reflecting the phase the organization is going through?