Job evaluation is not only a technical act.
A job evaluation always ends up producing a level, a score, a grade. And there is a stubborn belief: this result finalises the work. The method has run its course, the result is out, the function is positioned.
The technical result is only the first step.
A technical evaluation delivers a hypothesis. Solid, documented, but a hypothesis. What turns it into a defensible positioning is everything that comes after: bringing it into coherence, verifying it, validating it. It is this work that distinguishes a grade one endures from a level one can explain and trace.
In fact, it all starts before the method, with a rigorous reading of the function: what it does, what it does not do, who it interacts with, what it brings. This is a subject in its own right, to which I have devoted another article in this series, "Job description: reading what is written, and what is not". Without this reading, you evaluate a label. But let us assume this reading has been done. Here is what happens next.
A function is not evaluated in isolation
The score of a role taken on its own means nothing; what matters is its relative place within the whole.
In my practice, I build the evaluation from the top down, starting from the executive committee and working down to the rest of the organisation. The coherence of the levels is set at this stage, and this backbone becomes the reference that allows all other functions to be positioned.
This is a method in its own right, to which I have devoted another article in this series, "The top-down approach (building the evaluation end to end)".
Evaluating function by function, without an overall view, without setting the coherence of the first levels, produces a collection of scores that are individually correct and collectively inconsistent.
Verifying through cross-checks
Once the technical analysis has been done and the internal controls of the chosen method completed, the result must be confronted with the reality of the organisation. This is the cross-check stage. The positioning of a role is not read from a single angle, it is tested from several.
Its reporting line first: is the level of the role consistent with that of its manager, and with that of its own teams? An abnormal gap on this vertical is a signal. Its peers next: how does the role compare with other functions carrying a similar title elsewhere in the organisation? Its scope counterparts finally: how does it compare with roles in other business lines that have a comparable perimeter, or a comparable level of complexity or responsibility?
If the score holds up to these cross-checks, it is solid. If it does not, it is not the reality of the organisation that should be forced, it is the whole evaluation that should be reworked.
Validating with the business concerned
Then comes a step that some consider superfluous, and that I hold to be essential.
An evaluation result, however technically irreproachable and verified through cross-checks, must not remain HR's result. It must be validated with the HR business partner and with the business concerned. Not to be renegotiated, but to be tested against the view of those who live the organisation day to day, and to be carried by them.
This step has a virtue that is underestimated. An evaluation involves an irreducible human part. Whatever the method, it is a person who applies it, and that person has a hand: somewhat generous, somewhat strict. Acknowledging this is not an admission of weakness, it is a matter of humility and of respect towards the employees concerned. This evaluator bias is corrected in two ways. First, because the same evaluator handles the entire set of functions: their hand, whatever it may be, remains constant, and this constancy preserves what matters most, internal coherence. Second, because validation with the BP and the business brings different perspectives together and reduces the blind spot of an evaluation made by a single person.
Validation also has another virtue, more crucial still: checking the shared understanding of the function. In complex organisations, or for abstract or highly technical functions, it is the ultimate moment for verifying the evaluator's understanding, and the clarity of the information conveyed beforehand. It is also the moment to complete the information, orally during the exchange, and to formalise it afterwards in writing for traceability. The evaluator then distinguishes between what is genuinely new and adds to the rereading, and what is sometimes raised to inflate the function and try to obtain more points.
With clear and reasoned communication on the technique used, the checks carried out and the verification of internal coherence, it is rare to have to escalate to higher levels or to a validation committee to settle differences of view.
Evaluations rest on internal coherence like the mesh of a net, and a leader rarely wishes to unpick this coherence once it has been explained.
An evaluation verified and then validated is no longer "what HR decided". It becomes an organisational decision.
And what about external comparison?
There remains a question one sometimes asks: how do we compare with the market?
While job evaluations bring undeniable value to the organisation, comparing an internal level to a level known elsewhere can be misleading, and I would recommend a degree of caution, or at least an awareness of the possible biases. Three factors are likely to skew the exercise.
The real content of the function first. Take the example of a risk manager. Under an identical title, two roles can cover radically different realities: the precise description of the function, its exact scope, its metrics, the size of the company, its purpose (consulting firm or industrial production company), the structure of the organisation and the reporting lines (three levels in one organisation versus five in another, between the function and the CRO), how many similar functions exist in the company, and the internal coherence in which this role sits. So many variables that two risk managers are not necessarily comparable across two different companies, even if by chance they hold the same level.
The human part of the evaluator next, already mentioned: we do not know the hand that positioned the role elsewhere.
The level we are comparing to, finally, may have been pushed upwards by a promotion that was not really one, with no real leap in content. This drift of levels over time deserves attention: I will devote another article in this series to it, "The life of levels".
These biases open the door to a classic opportunistic move: invoking a level practised elsewhere to try to obtain an internal re-grading. "Everywhere else, my function is positioned one level higher." The argument, put forward without verification, is sometimes enough in companies with weak or insufficient structure.
The essentials
Whatever methodology is applied, the stake of a job evaluation is not the number obtained. It is the ability to compare, to explain and to trace a positioning that makes sense within the organisation.
A grade that cannot be explained does not hold. A level built methodologically, cross-checked, validated with the business and rooted in the reality of the organisation becomes a genuine governance tool.
This is the conviction that guides the HDH (HR Decision Hub) methodology: supporting companies in evaluating their functions, while keeping the traceability of decisions end to end.
www.hrdecisionhub.eu
#JobEvaluation #CompAndBen #PayTransparency #HDH