The seven dimensions of AI readiness, explained
9 min read · Updated 1 June 2026
AI readiness is easiest to act on when it is broken into dimensions. Below is a plain-language breakdown of seven, with what strong and weak look like in each, and the kind of evidence that distinguishes them.
1. Strategy and leadership
Strong: a named senior owner, AI goals tied to commercial outcomes, and committed (not just experimental) budget. Weak: AI is "someone's side project" with no link to revenue or cost. Evidence: a written objective, an owner on the org chart, a line in the budget.
2. Data foundations
Strong: data is accessible, reasonably clean, and owned. Weak: data is trapped in spreadsheets and inboxes, with no clear owner. Evidence: can a non-technical team pull an accurate dataset themselves, quickly?
3. Technology and infrastructure
Strong: systems expose data through modern interfaces and integrate with new tools. Weak: legacy systems with no integration path. Evidence: the presence of APIs, a cloud footprint, and the ability to connect tools without a rebuild.
4. Skills and culture
Strong: staff use AI tools confidently and leaders can separate real use cases from hype. Weak: fear, blanket bans, or uncritical enthusiasm. Evidence: training offered, tools sanctioned, and examples of staff improving their own work with AI.
5. Process integration
Strong: AI is embedded in a live workflow with a human-in-the-loop where it matters. Weak: permanent pilots that never reach production. Evidence: a process that genuinely depends on the model day to day.
6. Governance and risk
Strong: clear acceptable-use rules, UK GDPR compliance, and accountability for outputs. Weak: no policy, with sensitive data pasted into public tools. Evidence: a written AI use policy and a data-protection assessment.
7. Measurement
Strong: success is defined before a project starts and tracked afterwards. Weak: "it feels faster" with no baseline. Evidence: a metric, a before figure, and an after figure.
Reading your profile
The shape of your scores matters more than the average. A business that is strong on strategy but weak on data has a sequencing problem; one that is strong on data but weak on governance has a risk problem. Treat the lowest foundational dimension as the constraint that limits everything built on top of it.
Frequently asked questions
What are the dimensions of AI readiness?
A robust assessment covers strategy and leadership, data foundations, technology and infrastructure, skills and culture, process integration, governance and risk, and measurement.
Which AI readiness dimension matters most?
The lowest-scoring foundational dimension — typically data or governance — acts as the constraint on everything built on top of it, so it usually matters most to fix first.
What evidence shows strong data foundations?
The clearest signal is whether a non-technical team can pull an accurate, useful dataset themselves and quickly, without waiting on a specialist.