01 · 7 checks
Business case and sponsorship
Whether a named executive owns the outcome, the value hypothesis is quantified, and funding survives the first quarter without a fresh business case.
Free download
52 checks that tell you if you are actually ready for AI.
Seven dimensions, evidence-based questions and scoring bands that end with a recommended next move — foundations, pilot, scale or optimise. Written from delivery experience, not from a template.
What you get
A working document, not a download-and-forget PDF.
Fifty-two evidence-based checks across seven dimensions, with scoring bands that tell you whether to fix foundations, pilot, scale or optimise.
- Business case and sponsorship — 7 checks
- Data foundations — 8 checks
- Process readiness — 6 checks
- Technology and architecture — 8 checks
- Governance, risk and compliance — 9 checks
- People, skills and adoption — 7 checks
- Operating and scaling — 7 checks
- Scoring bands with a recommended next move for each
PDF · 3 pages · free
Written for: Technology, data, operations and risk leaders assessing AI maturity
The seven dimensions
Where AI programmes actually stall.
Each dimension carries its own checks and its own score, so you can see whether the blocker is data, governance or adoption rather than treating readiness as a single number.
02 · 8 checks
Data foundations
Source-system access, lineage, quality thresholds, master data, retention rules and whether the data an AI use case needs is reachable without a six-month project.
03 · 6 checks
Process readiness
How well the target process is documented, instrumented and stable enough that automation improves it rather than industrialising an existing mess.
04 · 8 checks
Technology and architecture
Environments, integration patterns, model hosting, evaluation tooling, observability and the route from a working prototype to a supported production service.
05 · 9 checks
Governance, risk and compliance
Model inventory, approval gates, human-in-the-loop rules, DPDP and sector obligations, audit evidence and incident handling for AI-specific failure modes.
06 · 7 checks
People, skills and adoption
Whether the people who must use the output were involved early, what training exists, and how adoption is measured beyond a launch-week usage spike.
07 · 7 checks
Operating and scaling
Run costs, ownership after go-live, retraining and drift monitoring, and the mechanism by which a second and third use case reuse the first one's platform.
Scoring bands
Every score ends in a recommended next move.
The point of scoring is not the number. It is knowing whether your next rupee belongs in foundations, a pilot, a scaling platform or optimisation.
0–25 — Foundations
Fix data access and pick one measurable process before committing to any AI build.
26–50 — Pilot
Run a tightly scoped pilot with an evaluation harness and a pre-agreed kill criterion.
51–75 — Scale
Industrialise the platform: shared evaluation, observability, governance gates and reuse.
76–100 — Optimise
Shift to portfolio economics — unit cost, model lifecycle and second-order value capture.
How teams use it
Three ways it earns its ninety minutes.
Board and steering-committee prep
Turn a broad 'what about AI?' question into seven scored dimensions and a defensible next move, so the discussion is about sequencing rather than sentiment.
Use-case triage
Score each candidate use case separately. The dimensions expose which idea is blocked on data, which is blocked on governance, and which is genuinely shovel-ready.
Vendor and partner selection
Take the checklist into supplier conversations. It gives you a consistent set of questions and makes weak answers visible across proposals.
Prefer an instant score?
Run the interactive assessment in ten minutes.
Our online AI readiness assessment scores you across six dimensions immediately, with strengths, gaps and next moves. Use it first, then use this checklist for the detailed evidence review.
AI readiness checklist — questions we get asked
It is a structured set of checks that tests whether an organisation can deliver and sustain AI in production — not just experiment with it. Ours covers 52 checks across business case, data, process, technology, governance, people and operations, and converts the answers into a score band with a recommended next move.
Most teams complete it in 45 to 90 minutes. It works best as a short working session with someone from technology, data, operations and risk in the room, because the disagreements between those four views are usually the most useful output.
Yes. The PDF is free. We ask for a work email so we can send the file and, if you want it, offer to walk through your scores. There is no obligation and you can unsubscribe at any time.
The online assessment is a 24-question interactive tool that scores you instantly across six dimensions. The PDF checklist is deeper — 52 checks across seven dimensions — and is designed for offline group work and evidence gathering. Many teams run the online assessment first, then use the PDF for the detailed review.
Yes. Bring your completed checklist to a discovery workshop and we will score it with you, size the two or three highest-value gaps and outline a first delivery increment with indicative effort.
Usually the transformation lead, CIO or head of data. The scoring is most useful when one person owns the consolidated view and each dimension has a named respondent who can produce evidence for their answers.
Every engagement starts with a
45-minute strategy session.
Bring the problem, the constraints and the deadline. You leave with an architecture opinion, a delivery shape and a realistic budget range — no pitch deck.