Maturity Model
The Framework Architecture gives you the structural picture — what each track owns, how the tracks depend on one another, and how to sequence them. The maturity model is the measurement layer on top of it: a single five-level ladder applied to all eight tracks, so you can score where the organization stands today and see the whole board on one comparable scale.
Maturity is assessed per track on the same five-level ladder defined in the AI Readiness Assessment Framework, so scores are comparable across the whole program:
1 Nascent · 2 Developing · 3 Emerging · 4 Scaling · 5 Transformational
The matrix below shows the anchor levels (1, 3, 5) for each track. Score on evidence, not intent — what is demonstrably true today, not what is planned.
| Track | Level 1 — Nascent | Level 3 — Emerging | Level 5 — Transformational |
|---|---|---|---|
| 01 · AI Strategy & Leadership | AI is a collection of experiments with no through-line | A funded AI strategy tied to a few business outcomes | AI strategy is inseparable from corporate strategy; capital reallocates on evidence |
| 02 · AI Governance & Risk | No policy; risk handled ad hoc per project | Acceptable-use policy and a review path exist for new use cases | Governance is automated and continuous; compliance is a byproduct of the platform |
| 03 · Data Readiness | Data is siloed and built for reporting, not AI | Key domains are governed and accessible to AI use cases | Data products are AI-ready by default, with lineage and contracts |
| 04 · Technology Architecture & Platform | Point solutions bought per team; no standards | A shared platform with standard model access and patterns exists | A self-serve internal platform; new use cases launch on paved roads |
| 05 · Workflow Optimization & Automation | AI used as scattered personal assistance | A few end-to-end workflows redesigned and in production | Agentic and redesigned workflows are the default operating model |
| 06 · AI Adoption & Culture | Tools rolled out; usage is low and shallow | Target teams use AI in daily work, with visible wins | AI-first thinking is the cultural norm; people redesign their own work |
| 07 · Talent & Capability Building | AI literacy is rare and individual | Role-based enablement and internal champions exist | Capability building is continuous; roles are designed around human–AI collaboration |
| 08 · Measurement & Value Realization | Value is asserted, not measured | Key initiatives are instrumented and attributed | Measurement closes the loop and actively reprioritizes investment |
Two patterns to watch for in a completed assessment:
- Flat-but-low (everything at 2) usually means no track has reached the threshold where value appears — pick one or two and push them to Level 3–4 rather than nudging all eight.
- Spiky (a 4 next to a 1) signals a binding constraint: the high track is being throttled by the low one. The Data Readiness → Technology Architecture & Platform boundary, and the boundary from Workflow Optimization & Automation into AI Adoption & Culture and Talent & Capability Building, are the most common places to stall — working pilots that can't scale almost always trace to a Level-1 dependency next door.
To turn these scores into a prioritized investment case — how to source each track's score, reconcile it in a cross-functional working session, and read the full set of profile shapes (spiky, flat-but-low, foundation-gap) — see the Integrated Assessment, the how-to-run-it companion to this model. To score your organization interactively, use the AI Maturity Self-Assessment.