AI Readiness Score
Tick what's true for your organisation. Your live score and tailored next step update as you go.
How the AI readiness score works
A tool that will not show its arithmetic is worth less than one that does.
- 1
You answer questions across data, infrastructure, skills, governance and executive sponsorship.
- 2
Each dimension is scored separately rather than averaged into one figure, because a high average hides the dimension that will actually stop the project.
- 3
The output is a per-dimension profile with the weakest area called out first.
What it assumes
Self-assessment. Teams consistently overrate their data quality and underrate the governance work, so scores tend to run optimistic.
The dimensions are treated as independent, though in practice weak data quality makes every other dimension harder.
It measures readiness to deploy, not whether a given use case is worth doing. Those are different questions.
How to read the result
Read the lowest dimension, not the total. AI projects fail on their weakest constraint, and it is very often data access or ownership rather than modelling capability. A team scoring well on skills and badly on governance should fix governance before writing code — the alternative is a working prototype that cannot be put in front of a customer.
Questions
- What score means we are ready?
- There is no threshold, and a tool that gave you one would be misleading you. What matters is whether your weakest dimension is strong enough for the specific thing you want to ship. A narrow internal pilot needs far less than a customer-facing deployment.
- Which dimension blocks projects most often?
- Data access. Not data quality or volume — access. The data usually exists; the delay is agreeing who owns it, who may use it and under what terms.