Most organizations that ask us about AI do not have an AI problem. They have an unanswered question: if we started tomorrow, what would stop us?
An AI readiness assessment answers that question. It scores the separate conditions an AI project needs to reach production, such as data access, team skills, and the systems the AI must connect to, and it names the weakest one. That weakest condition is your binding constraint: the one limit that will stop a project first, whatever else is in place. Find it before a project budget finds it for you.
What an AI readiness assessment should measure
A useful readiness assessment scores each condition separately, because a strong score in one area hides a weak score in another. Plenty of assessments are twenty-page questionnaires that end in a percentile and a sales call. The useful kind is narrower. It scores only the dimensions that decide whether an AI project ships.
Our free assessment scores six: AI usage, data readiness, team skills, investment, organization, and tech stack. It takes about two minutes. Each dimension gets its own score on a radar chart, next to the overall average.
Data readiness deserves the closest look. In a February 26, 2025 press release, Gartner predicted that through 2026, organizations will abandon 60% of AI projects unsupported by AI-ready data. The same release reported that 63% of organizations in its Q3 2024 survey of 248 data management leaders either do not have the right data management practices for AI or are unsure whether they do.
Why the lowest score matters more than the average
The lowest dimension score is the result to act on, because it marks where your first dollar should go. The average tells you how you compare. The lowest score tells you what will break.
The failure pattern we see most often is an organization that is strong in one dimension and assumes the strength carries over. Executive sponsorship, with data nobody can get out of the system that holds it. Clean data, with no engineer who can connect a model to the application that produces it. A capable team, pointed at a system with no APIs. Each of those organizations would post a respectable average. Each would stall on the first real build.
What to do with each result
The overall score places you in one of three stages, and each stage calls for different work.
Exploring (2.0 or below). Do strategy work before any pilot. A use case picked before you understand your constraints becomes a demo that cannot leave the demo environment.
Building (2.1 to 3.5). The foundation exists, so the risk moves to execution. Pick the one workflow that justifies a production build, and run the pilot on real data from the first day. Most stalled pilots sit at this stage. Our guide to running an AI pilot that produces a decision covers how to scope one.
Scaling (above 3.5). The assessment mostly confirms what you know. The work is governance, training, and spreading what already works to the teams that have not seen it. For the governance part, the NIST AI Risk Management Framework, a voluntary framework NIST released on January 26, 2023, is a sensible anchor.
At every stage, check the lowest dimension before you act on the stage. An organization in Building with data readiness at the floor is still Exploring for any project that depends on that data.
Where the assessment fits in an engagement
The two-minute assessment gives you a baseline, not a plan. We use it as the entry point to the Assess phase of our AI implementation framework. That phase audits the same areas in depth (data quality and access, workflows, infrastructure, and team readiness) and ends in a prioritized roadmap instead of a score.
If you plan to bring in outside help for that step, keep our questions to ask an AI consultant before you sign next to your result. A consultant who does not ask about your lowest score is guessing.
If the result surprises you, the surprise is the most useful thing the assessment produced. It means the constraint you were about to spend money on was not the real one. Take the AI readiness assessment and see where you actually stand.