Evaluating the functions of an organisation is not handling them one after another in a long list. It is building a structure, one that is built in a precise order. In my practice, this order goes from the top down.
Why start from the top?
The level of a function has no meaning in isolation. It only has meaning in relation to the others. If you start from the bottom, or from the middle, you position roles without yet having the framework that allows you to judge that positioning. You accumulate evaluations that are individually correct, and shaky together.
Starting from the top means first setting the most structuring reference points, the ones on which everything else will depend. It means defining the working framework.
How the sequence unfolds?
The sequence is not arbitrary, it follows the logic of the anchor points.
You first position the CEO, according to the methodology chosen by the organisation, whether analytical by points or by grades. This first positioning sets the ceiling of the structure, the upper reference on which everything else will depend.
Next comes a major player, finance. Finance often sits one or two levels below the CEO, and this elevated position is not arbitrary: finance is intrinsic to the organisation.
It is a privileged partner of the business, and takes part concretely in steering and in all strategic decisions. It is also the centre of merger and acquisition operations, and its scope extends to the entire asset base of the organisation, from buildings to budgets, from provisions to investments, all the way to the most unexpected items on the balance sheet. Positioning finance early therefore sets a second major anchor point, almost as structuring as the first.
Right after comes the business itself. What does the company produce, services, products, both? And above all, what is its strongest business line?
This is not a question of headcount, but of Net Revenue or EBITDA: the business line that is economically the most significant will serve as the reference for positioning the others.
Its positioning is refined by crossing several elements: its reporting line, the impact measured by the weight of what is handled, and the other criteria specific to the chosen methodology, such as responsibilities or complexity.
Then come the leaders of the departments that collaborate with each of these business lines. Then the support functions of those same business lines.
At each step, the framework already set serves as the reference for the next step. You never position a function in a vacuum: you position it in relation to what has already been established above it and beside it.
The moment to bring the leaders together
Once this first construction is done, comes a decisive step. You gather in the same room the CHRO, the members of the executive committee and their direct N-1s, and you discuss the positioning together.
This is not a validation formality. It is the moment when perspectives meet, when inconsistencies surface, when trade-offs are made aloud rather than in the silence of a spreadsheet.
It is also the moment for a detailed description of the chosen methodology and of its direct application, the moment of buy-in by the extended executive committee.
It is also, to my mind, the moment when external support adds the most value. If an organisation had to choose a single phase to bring in external support, this would be the one. An outside perspective helps to avoid starting on the wrong foot, and to defuse the internal tensions that can crystallise around relative positionings. Once the reference structure is well set, the rest of the work can be carried out autonomously.
The moment to set governance
This same meeting is also the right moment to decide on governance. Do you set up a level review committee? Which levels will be validated, and by whom?
The answers belong to each organisation. The group of N-1s, accompanied by the evaluator and the relevant HR business partner, may for example validate the N-2 and N-3 levels, while the other functions are validated directly by the evaluator, the HR business partner and the direct manager, with an escalation option. These are only examples: what matters is that these rules are set consciously, and at the right moment.
Keeping this governance alive over time is a subject in its own right, which I address in another article in this series, "The life of levels". What matters here is to remember that governance is decided at the moment the structure is built.
The essentials
A solid job architecture does not come from an addition of functions evaluated separately. It comes from an ordered construction, from the top down, where each level set becomes the reference for the next.
It is at the moment when this structure is being built that its governance is also decided: who validates which levels, and under which rules. Keeping this governance alive over time is a subject in its own right, which I address in another article in this series, "The life of levels".
This is the logic of structured construction that the HDH (HR Decision Hub) platform recommends, while keeping the trace of decisions and of what justifies them at every stage.