The ten AI readiness dimensions, named and explained

We wrote last time about what a readiness score actually represents: a single figure compressed from measurable dimensions of current practice, useful for a board pack precisely because it hides the detail underneath. This piece opens the box.

None of what follows is a secret formula. It is a plain list of what the ten dimensions are, in language that does not require the assessment tool itself to understand.

Why ten, and why these ten

Fewer dimensions and the score stops distinguishing between organisations that are genuinely different. More, and nobody outside the assessment team remembers what is actually being measured, which defeats the point of naming them at all. Ten is the number that has, in practice, held up across the organisations we have assessed: enough to separate operational competence from organisational climate, few enough that a board member can hold the list in their head after one reading.

The ten dimensions, named and explained

• Fluency. Basic operational competence: can staff use the tools available to them at all, without needing a colleague to drive.
• Delegation practice. Whether staff know which tasks are safe to hand over and which are not, the judgement we have described elsewhere as reading the jagged frontier.
• Description discipline. Whether a person can specify a task precisely enough that a tool, or another colleague, executes it correctly the first time.
• Verification. Whether outputs get checked before they are used, and by whom, rather than accepted because they read fluently.
• Logging. Whether AI-assisted work is recorded anywhere at all, so it can be reconstructed six months later when someone asks how a decision was reached.
• Data protection. Whether the classification and handling rules an organisation has written down are the rules staff actually follow under deadline pressure.
• Workforce impact. How the content of people's jobs is actually changing: what is disappearing, what is intensifying, and whether anyone is tracking it.
• Internal politics. Who gains standing from adoption, who loses it, and whether that quiet calculation is accelerating or blocking uptake regardless of the official policy.
• Psychological safety. Whether staff feel able to say they do not know how to use a tool, or that they made a mistake with one, without reputational cost.
• Future-readiness. Whether the organisation is positioned to keep adapting as the tools change, rather than optimised for the specific tools in use today.

How the ten cluster

The first six sit closer to operational practice: fluency, delegation, description, verification, logging, data protection. These move with training, tooling, and clear procedure, and they tend to respond within a quarter to direct intervention.

The last four sit closer to organisational climate: workforce impact, internal politics, psychological safety, future-readiness. These do not move with a training session. They move with how leadership behaves under pressure, and they are the dimensions most readiness assessments quietly under-weight because they are harder to measure and less comfortable to discuss.

A dimension nobody can name is a dimension nobody is accountable for.

That is the practical case for naming all ten in plain text rather than leaving them inside a scoring tool: a dimension with a name can be assigned an owner, tracked, and argued about. A dimension that only exists as a slider inside a calculator cannot.

The practical takeaway

Before your next readiness conversation, ask which of the ten dimensions has an owner and which does not; an unowned dimension is not being managed, whatever the top-line score suggests.

No external statistic cited; this article defines the organisation's own assessment methodology rather than presenting third-party evidence.