Key takeaways
- AI projects fail more often from unclear goals and poor data than from weak models.
- Seven questions cover the essentials: value, data, process, technology, people, risk and measurement.
- Readiness gaps are normal — the goal is to know them before you invest.
- The output should be a short, prioritised roadmap, not a long report.
Every leadership team is being asked, “What is our AI plan?” Jumping straight to tools often leads to pilots that never reach production. A short AI readiness assessment answers a better question first: where can AI create real value for us, and what do we need in place to deliver it?
Question 1: What business outcome are we targeting?
Start with a measurable goal — faster response times, lower processing cost, higher conversion, fewer errors. “Use AI” is not a goal. Tie each idea to an owner and a metric.
Question 2: Is the data available, accessible and good enough?
- Where does the data live, and can systems access it through APIs?
- Is it reasonably complete, consistent and current?
- Are there legal or contractual limits on how it can be used?
Question 3: Is the process understood and stable?
If nobody can describe how the work is done today, AI will not fix it. Map the steps, decisions and exceptions first — see our guide to business process automation with AI.
Question 4: Can our technology support it?
Check identity and access management, integration options, cloud readiness and security tooling. Legacy systems are not a blocker, but they shape the approach.
Question 5: Do we have the people and sponsorship?
- An executive sponsor who owns the outcome.
- A business owner for each use case.
- Technical capability — internal, partner or both.
- A plan to train and support the people whose work will change.
Question 6: How will we manage risk?
Define acceptable use, data handling, human oversight and incident response before the first pilot. Our responsible AI governance checklist is a practical starting point.
Question 7: How will we measure success?
Capture baselines now. Decide which metrics prove value, how often you will review them and what result would justify scaling.
Turning answers into a roadmap
| Horizon | Focus |
|---|---|
| Next 90 days | One or two quick-win pilots with clear metrics; fix critical data gaps |
| 3–6 months | Move successful pilots to production; set up governance and monitoring |
| 6–12 months | Scale across departments; build shared platforms and skills |
These horizons are typical planning estimates; actual timing depends on your systems and priorities.
Get a clear, practical AI roadmap
Our AI & Digital Transformation team runs focused readiness assessments and turns them into prioritised plans.
Frequently asked questions
What is an AI readiness assessment?
It is a structured review of your goals, data, processes, technology, people and governance to identify which AI opportunities are achievable and what must be improved first.
How long does an AI readiness assessment take?
A focused assessment is usually planned over a few weeks, depending on the number of departments and systems involved.
Do small and mid-sized businesses need one?
Yes, in a lighter form. Even a short workshop helps smaller teams avoid investing in tools that do not match their needs.
What do we receive at the end?
Typically a prioritised list of use cases, readiness gaps, risks and a phased roadmap with suggested pilots and success measures.
