Why a maturity model helps
A maturity model is not a report card. It is a map. It tells you where you are and what the next move looks like. Without a map, AI investment becomes a series of disconnected experiments and the company never builds compounding capability.
This model has five levels. Each one has a clear definition and a clear next step.
Level 1: AI curious
Symptoms: people are using ChatGPT and other tools individually. No formal strategy. No governance. No measurement. No CEO-level sponsor.
What to do: name an executive sponsor. Run a Fractional CAIO pilot. Pick three use cases and ship at least one to production within 90 days.
Level 2: AI active
Symptoms: one or two pilots in flight. Some shadow IT. A growing list of AI vendors. No governance. No measurement framework.
What to do: kill the projects without measurement. Build a governance posture. Lock the pilots into a real prioritization framework. Prove ROI on at least one production use case.
Level 3: AI aligned
Symptoms: a defined AI strategy with executive ownership. Two to five production use cases. A working governance posture. Measurement in place. Vendor consolidation underway.
What to do: scale the portfolio. Build the internal team. Start moving from internal leverage use cases to customer-facing ones. Define the long-term operating model.
Level 4: AI scaling
Symptoms: AI is part of the operating cadence. The board sees AI impact every quarter. Multiple business units are running their own use cases inside a shared framework. The CoE is functional.
What to do: tighten the governance posture as the surface grows. Prepare for a full-time AI leader if not already in place. Begin AI-native product or service development.
Level 5: AI native
Symptoms: AI is embedded in the operating model and the product strategy. New initiatives are scoped AI-first. Talent retention in AI roles is high. The company is a credible AI employer in the market.
What to do: continue compounding. The work shifts from building capability to maintaining strategic edge. This is where the competitive moat lives.
Where most mid-market companies sit
There is no census of mid-market maturity, so treat this as a judgment backed by public surveys. In McKinsey's November 2025 state of AI survey, roughly one third of respondents said their companies had begun to scale AI. BCG's September 30, 2025 study of more than 1,250 companies found about 60 percent reporting little or no value from AI, and about 5 percent achieving value at scale. Sitting at level 1 or level 2 is common. The gap is opportunity, not embarrassment, and a Fractional CAIO engagement is one way to shorten the climb.
A five-question self-check
A rule of thumb, not a scored instrument. Answer yes or no to each question.
- Does a named executive own AI outcomes?
- Is there a written governance posture that people actually follow?
- Is at least one use case in production with real users?
- Is its impact measured in revenue, margin or cost terms?
- Does AI appear in the regular operating and board cadence?
No or one yes usually places a company at level 1 or 2, and the next move is a sponsor, governance and one measured use case, as above. Two or three yes answers point to level 3 and a question of scaling the portfolio. Four or five point to levels 4 and 5, where the work shifts to tightening governance and protecting the edge. For a calendar to act on the result, see the 90 day roadmap.
Sources
- McKinsey, The state of AI in 2025: Agents, innovation, and transformation (November 2025). In a survey of 1,993 respondents, 88 percent reported using AI in at least one business function, roughly one third said their companies had begun to scale it, and 39 percent reported any enterprise-level EBIT impact.
- BCG, The Widening AI Value Gap (September 30, 2025). In a study of more than 1,250 companies, about 60 percent reported little or no value from AI, while about 5 percent were achieving value at scale.