A 5-Level Framework for Measuring Workforce Execution
Access to AI tools has spread faster than the ability to use them well. A professional in Lagos, Nairobi or Johannesburg can open the same model, on the same day, as a counterpart in London or Singapore. What differs is what happens next: whether that access becomes a documented workflow, a measurable output, a governance decision, or nothing at all.
The gap is not information. It is execution, and execution is shaped by local conditions: the systems a workplace already runs on, the visibility the market grants its professionals, and the resilience of the income behind the work.
There is also no shared instrument for the thing that now matters most. iMocha’s recent report on AI fluency in the workforce makes the point directly: AI awareness is now widespread, yet no widely adopted standard exists for measuring AI fluency. Without a standard, every institution defines capability differently, and comparison across cohorts, sectors and countries becomes guesswork.
ARETE is publishing this scale to give African institutions a common benchmark: five levels defined by what a person does rather than what they have heard, applied consistently to professionals, teams and cohorts.
Each level is defined by observable execution, not self-reported confidence. A person sits at the highest level whose description their working practice satisfies.
Understands AI concepts but does not use them.
AI is known as a subject. The vocabulary is familiar, the claims are understood, and the shift is followed closely. It does not yet appear in the working week: no recurring task is handed to a model, so nothing in the person's output bears the mark of it.
What moves a person up: Hand one recurring task to a model, deliberately, for a fortnight.
Experiments with AI tools but lacks consistent execution.
Tools are tried. A few are kept, most are dropped, and the result depends on available time rather than on process. Improvements arrive occasionally and unpredictably, and cannot be reproduced by anyone else, including the person who produced them.
What moves a person up: Fix one workflow so its AI step happens every time, the same way.
Uses AI effectively for daily tasks and individual workflows.
AI is part of the working day for individual tasks: drafting, analysis, research, preparation. Output is reliable and faster than it was. The limit is structural — the benefit stops at the person using it, and leaves the room with them.
What moves a person up: Document the workflow so a colleague can run it without asking.
Integrates AI into team processes, systems, and measurable outputs.
AI sits inside how a team works: in handoffs, in documented processes, in quality checks, and attached to outputs the organisation actually measures. Capability is no longer personal. It survives the individual and scales across a unit.
What moves a person up: Write the rules that govern the system: what it may decide, who is accountable, what is audited.
Designs AI governance, strategy, and systemic transformation.
The person or team designs the system rather than operating inside it: AI strategy, governance, risk tolerance, procurement, and the transformation agenda across the institution. This is where individual fluency becomes institutional authority.
What moves a person up: This is the ceiling of the scale. The work here is maintaining the standard and measuring the rest of the institution against it.
The scale is mapped to the 12-pillar ARETE Readiness Instrument, which measures execution across direction, systems, visibility, income resilience and AI capability rather than self-reported awareness. Every submission produces a score out of 100, and that score places the respondent on exactly one level.
The five levels sit inside the four readiness categories the assessment already reports; the split at the top separates professionals who apply AI inside existing systems from those who design and govern those systems.
The scale is in active use. Every professional and cohort assessed through ARETE is scored against these five levels, which makes it possible to report a distribution for one institution, one sector, or the continent as a whole. The published ARETE Readiness Index is built on the same instrument.
Want to know where you or your institution sits on this scale? Take the 12-pillar assessment.