What Levels of Work Change About Competency Models
Most competency models describe behaviours as if they were the same at every level of an organization. They are not. The Levels of Work lens explains why the same named competency is a different job at Level 2 than at Level 5.
A competency model that lists “strategic thinking” as a required behaviour, and then applies the same definition to a shift supervisor and a division president, has already failed. Not because the behaviour is wrong, but because the word means two different things in the two roles.
The problem with flat models
Flat competency models are attractive because they are portable. One list, one rating scale, one set of interview questions, applied across the whole organization. They are easy to administer and easy to defend. They are also close to useless for the decisions that matter most — who is ready for the next role, and what has to change before they are.
The failure mode is consistent. A leader is rated highly on a competency at their current level, promoted on the strength of that rating, and then struggles. Nothing about their character changed. What changed was the horizon of the decisions they now own.
What the lens adds
The Levels of Work lens sorts roles by the complexity and time span of the decisions attached to them, rather than by title, band, or headcount. Once a role is placed, a competency stops being an abstract behaviour and becomes a concrete question: at this horizon, with this degree of ambiguity, what does this competency actually require?
That question produces different answers at each level:
- At the first two levels, competence usually shows up as reliability and responsiveness inside a defined standard.
- In the middle, it becomes the ability to improve the standard itself, and to see the system that produces the results.
- At the top, it becomes the ability to choose what the organization will be good at, and to accept responsibility for the consequences of that choice over years rather than quarters.
Same word. Different work.
Why this matters for how you assess
Once you accept that competencies are level-dependent, three practical things follow.
Calibration has to be level-aware. Comparing two leaders on the same numeric rating is meaningless if they are operating at different horizons. The rating has to carry the level with it.
Development has to name the next horizon, not the next behaviour. Telling someone to “be more strategic” is not actionable. Telling them which class of decision they will own next, and what that decision requires, is.
Selection has to test at the target level. Evidence of strong performance at the current level is necessary but not sufficient. The question is whether the person can hold the ambiguity of the level above.
None of this makes competency models unnecessary. It makes them usable.
Where this sits in the book
The Leadership Competency Playbook organizes twenty-four competencies through the DOES cycle — Design, Organize, Execute, Sustain — and reads each one across the five Levels of Work. The second edition treats the levels lens as the spine of the book rather than an appendix to it.
