An AI readiness checklist: ten questions

You can get a rough read on your AI readiness in about fifteen minutes: answer ten yes-or-no questions, one for each sub-score in the assessment, and look at where the “no” and “not sure” answers cluster. This is a self-check, not a grade. It shows you where to look first.

Answer for your organization as it is today, not as you plan for it to be. If you have to guess, write “not sure.” That is an honest answer, and it counts as a finding.

Pillar 1: Information & Data Foundation

1.1 Data inventory and ownership. Can you name where your important data lives, and one person responsible for each place?

If the list lives in someone’s head, the answer is no.

1.2 Information architecture and findability. If you ask three people on different teams to find your current price list, do they come back with the same file in under two minutes?

Different files, or a long hunt, means an AI assistant will have the same trouble. Why structure comes before any AI tool.

1.3 Data quality and governance. Is there an agreed rule for which version of a document is the official one, and for retiring the old ones?

1.4 Security and access control. Do you know who can open your HR, finance, and legal files today, and is that list right?

An AI assistant does not add access. It makes existing access much easier to use. What Copilot can see.

Pillar 2: Business Direction

2.1 Strategic intent and use cases. Can you state, in one sentence with a number in it, the first problem you want AI to help with?

“We want to be more efficient” does not count. “Two estimators spend about ten hours a week re-typing supplier quotes” does. Name the problem before the tool.

2.2 Sponsorship and budget. Does a named leader own the effort, and is there money set aside for more than licenses, such as training and cleanup?

2.3 Measurement and return. Have you recorded today’s baseline for that problem, so you can tell later whether anything improved?

Pillar 3: People & Tool Readiness

3.1 Staff skills and AI literacy. Have your people had any training on what these tools do well, what they get wrong, and what not to paste into them?

3.2 Platform and tooling. Do you know which AI features your current licenses already include, and which tools your staff are using on their own?

3.3 Policy and change management. Is there a written acceptable-use policy, and has every team heard it explained? A single page is enough. What goes on the page.

How to read your answers

Do not add up a score. Look at the pattern instead.

• Mostly “no” in Pillar 1 means the first project should be structure and access, not a new tool.
• Mostly “no” in Pillar 2 means a short workshop to pick one problem, one owner, and one number will do more than any purchase.
• Mostly “no” in Pillar 3 means a one-page policy and thirty minutes of training per team, which are among the fastest gaps to close.
• “Not sure” scattered everywhere is the most common result, and it is normal. It means the facts have not been gathered yet.

Most organizations that take the assessment for the first time land between C+ and C-. A few “no” answers here are not a verdict. They are the list you start with.

What the assessment adds

The checklist runs on opinion. The assessment replaces opinion with evidence: we look at your actual sites, permissions, and policies, grade all ten sub-scores, and give you a roadmap in fix-first order. If you want to see what that looks like, you can download two sample scorecards, one for a company and one for a city or county. Both use fictional clients. When you are ready to see your own, read how the assessment works.