How I assess AI readiness
AI readiness is the organisation's ability to use AI responsibly and make it work in day to day operations. A useful assessment connects the business goal with the evidence, controls and capabilities needed to act.
I examine what is documented, what people describe and what can be seen in practice. The result is a set of priorities, owners and evidence gaps that helps leadership choose the next move.
This readable method summary accompanies the redacted visual sample. It does not reproduce client findings or claim that a score guarantees business value.
What does the assessment examine?
Data foundations. Data quality, sensitivity, access and ownership. Can the proposed use case use the data it needs?
Process and use case fit. The work being changed, its owner and the quality standard for an accepted outcome. Where does AI help and where must a person decide?
People and change. Skills, roles, training and the working habits needed for adoption.
Governance. Policies, approvals, accountability, customer obligations and controls that operate in practice.
Technology and tooling. The approved tools, integrations, operating responsibilities and controls around AI agents.
Strategy and ambition. The business priorities, target state and decisions that the organisation is ready to fund and support.
How do I judge the evidence?
A policy on paper and a control in everyday use are different evidence. Interviews explain the context. Documents, records and examples of actual work help test it. A correction confirmed by the relevant source replaces the earlier draft wording.
Readiness, confidence and coverage are recorded separately. Readiness describes the current capability. Confidence describes how strongly the evidence supports that judgement. Coverage shows how much of the requested evidence is available. Missing evidence remains visible.
What does a maturity score tell you?
The sample uses five levels: Ad hoc, Emerging, Defined, Managed and Optimized. A score describes the evidence at the time of assessment. The target is a future state supported by specific changes. Weightings, assumptions and evidence gaps belong beside the score so a reader can challenge the conclusion.
How are blockers and next steps handled?
A blocker is a gap that slows or stops safe use. The review considers likelihood and consequence, then identifies an owner and a practical test for closing the gap. Dependencies help establish the order of work. Some decisions can sit with an operating team; others need leadership approval.
The written output can include area findings, evidence gaps, a blocker register, decision points and a sequenced roadmap. The agreed scope determines the outputs. A readiness assessment does not by itself establish savings, investment returns or permission to deploy a particular AI system.
How does readiness differ from AI value?
Readiness asks whether the organisation can support the work. AI value asks whether a particular use case creates a worthwhile outcome after its full costs and limitations are considered. Keep these judgements alongside each other.
Read the AI value method or review the AI Readiness Assessment scope.