The five levels of AI readiness — and why most companies are at level two
By CTC Smart Tech · 8 September 2026 · 6 min read
Before choosing a use case, know where your data, processes and people actually are.
We assess AI readiness on five levels: (1) individual tools, (2) team experiments, (3) one process in production with measurement, (4) several processes with governance, (5) AI as an operating capability.
Most companies that call us are at level two: enthusiastic individuals, a few demos, no measured process. The temptation is to jump to an ambitious agent. The right move is one process at level three — with a baseline, an approval step and a quality score.
Level three is where the organisation learns what AI costs, what it breaks and how to govern it. Level four follows naturally; level five is rare and does not need to be the goal.
Our assessment produces the level, a backlog of use cases ranked by value and effort, and the pre-requisites for the first two. Sometimes the honest answer is: fix the data first.
