What’s Your AI Readiness Score? How the 0–100 Scale Works

“What’s our AI readiness score?” Leadership teams like a single number, and I understand why. It fits on a slide, it can be compared quarter to quarter, and it gives a board something concrete to ask about. But a single number is only useful if you know how it was built. Otherwise, a 55 is just a feeling with a decimal point.

This post explains exactly how the AI readiness score in the AI Readiness assessment is calculated: where the 0 to 100 comes from, what the four rating bands mean, and the one rule that can override the average. It uses the published sample report as a worked example, so you can check every number yourself.

Where the number comes from

The assessment asks 24 questions, four in each of six dimensions: data, security, infrastructure, skills, use cases, and governance. Every question has five possible answers, each describing a maturity level from 0 to 4. You pick the answer that honestly describes your organization today.

From there, the math is deliberately simple:

• A dimension’s score is the points earned divided by the points possible. With four questions worth up to 4 points each, a dimension has 16 points possible, so each point you earn is worth 6.25 points on that dimension’s 0 to 100 scale.
• The overall score is the plain average of the six dimension scores. No dimension is weighted more heavily than another in the average.

That’s the whole calculation. The score is deterministic: the same answers always produce the same number. No AI model decides it, and no consultant adjusts it.

A worked example

The sample report is written for a fictional mid-size manufacturer. Its dimension scores are:

• Data readiness: 38
• Security and privacy: 50
• Infrastructure and platforms: 50
• Skills and talent: 31
• Use cases and value: 38
• Governance and operating model: 25

Add them up and you get 232. Divide by six and you get 38.7, which rounds to 39. That places the company in the Not Ready band, even though two dimensions are at 50. The shape tells the real story: infrastructure is ahead of the guardrails, and governance is the weakest area by a clear margin.

The four rating bands

The overall score falls into one of four bands. Each has a short description in the report:

• Not Ready, 0 to 39: “Foundations are missing. Adopting AI now adds risk faster than value.”
• Emerging, 40 to 59: “Pieces are in place, but gaps will stall pilots or create exposure.”
• Ready, 60 to 79: “Solid foundations. Ready to run and scale targeted use cases.”
• Leading, 80 to 100: “Mature and governed. Focus on scaling value and building.”

A useful way to read the bands: an average answer of level 3 across a dimension produces a score of 75, inside Ready. An average of level 2 produces 50, inside Emerging. So “Ready” roughly means most of your answers sit at level 3, where things are owned, documented, and reviewed rather than done ad hoc.

The floor rule

There’s one exception to the plain average. If security or governance scores below 40, the overall rating is capped at Emerging, regardless of the average. An organization with excellent infrastructure and skills can’t be rated Ready if it’s leaking data or has no way to decide what’s allowed.

The rule exists because those two dimensions carry risk that the others can’t offset. Strong skills don’t reduce the exposure from overshared documents; a mature cloud platform doesn’t substitute for an acceptable use policy. If you’re in this situation, the fastest way to raise your rating is to work on those two dimensions first. AI security readiness and AI governance for mid-size IT cover where to start.

Why a deterministic score matters

It might seem more sophisticated to have an expert or an AI model weigh everything and produce a judgment. In practice, a fixed calculation is more useful for three reasons. It’s comparable over time: if your score rises next quarter, it’s because your answers changed, not because someone read them differently. It’s comparable across teams: two business units scored the same way can be compared fairly. And it’s defensible: when a board member asks why the score is 39, you can show them the six numbers and the average. The judgment belongs in the recommendations, where your context matters. The number itself should be something anyone can check.

What the score doesn’t tell you

A number can tell you where you stand. It can’t tell you what to do about it in your specific situation. Two organizations with the same score can need very different plans, depending on their industry, their regulations, their main platform, and what they want AI to do in the next year. In the full report, the score comes with risks ranked from your answers, quick wins, and a 90-day plan written against your context. The number is the starting point, not the conclusion.

It also doesn’t capture red flags on its own. Certain answers, such as broadly overshared content or unmanaged use of public AI tools, go to the top of the risk list whatever the overall score is.

The asset: what to do in each band

• Not Ready: fix any red flags first. Publish an acceptable use policy and give people an approved AI tool. Run only low-risk pilots with a single team, and hold off on broad rollouts and custom builds.
• Emerging: run one or two measured pilots with business owners and baselines. Put most of your effort into your weakest dimension.
• Ready: scale the use cases that have proven value, start the second and third, and formalize the governance cadence.
• Leading: focus on scaling value, building your own solutions, and adopting a formal framework if customers or regulators expect one.

How fast can the score move?

Because the math is simple, you can predict the effect of improvements. Moving one question up one level adds 6.25 points to that dimension and a little over 1 point to the overall score. Moving all four questions in a dimension up one level adds 25 points to the dimension and about 4 points overall. That’s why focusing on your weakest dimension usually moves the overall score fastest: its questions tend to have the most room to climb, and the improvements compound in one place.

Re-score every quarter. The level above your current answer on each question is a reasonable goal for the next quarter, and watching the dimension scores change is more motivating than watching the overall number. Keep each quarter’s answers alongside the scores, so you can see exactly which questions moved.

Scoring yourself

You can calculate your own score from the AI readiness checklist, which lists all 24 questions, and how to measure AI readiness without a consulting engagement walks through running the exercise with your team. To see how the score fits into a full report, what a real AI readiness report looks like walks through the sample page by page.

Where does your team actually stand?

The free AI Readiness Score uses 10 of the 24 questions, spread across all six dimensions, and gives you a score in a few minutes.

Get your free AI Readiness Score →

Want to see the score in context first? Flip through a complete 38-page sample report.

Related guides

• AI Security Readiness: Closing the Gaps AI Exposes
• AI Governance for Mid-Size IT: The Lightweight Operating Model
• The AI Readiness Checklist: 24 Questions to Answer Before Spending a Dollar
• How to Measure AI Readiness (Without a Consulting Engagement)
• What a Real AI Readiness Report Looks Like (38-Page Example)