Frameworks
the readiness scorecard, the maturity scale, the four problem shapes
The instruments, published
These are the actual scoring instruments the audit uses. They are here so a buyer can judge them before paying for one.
A readiness score with no benchmark behind it is meaningless, and most published assessments do not say which they are. So the discipline that matters is disclosure: where a real external benchmark exists for a dimension we state it, and where none does we say so rather than implying a precision we do not have.
Six readiness dimensions
Readiness is not a generic organisational property. It is readiness for something, and the audit names that something first.
A company can be entirely ready for document classification and entirely unready for predictive pricing. So each dimension is scored one to five against two or three named candidate use cases, with the evidence behind every score recorded - not scored in the abstract, which produces a number nobody can act on.
The six dimensions, scored one to five. The shape shown is a worked example, not a client's score.
- 1 Strategy & Vision Alignment 3
- 2 Data Infrastructure & Quality 2
- 3 Technology & Architecture 3
- 4 Talent & Skills 2
- 5 Governance & Ethics 2
- 6 Research/Benchmarking Capability 1
1 = unaware · 5 = transformational
Default weights, and how they are adjusted for this region
| Dimension | What the score is evidenced against |
|---|---|
| 1. Strategy & Vision Alignment | Named exec sponsor? Roadmap with resourcing? |
| 2. Data Infrastructure & Quality | Accessibility, governance, fitness-for-purpose |
| 3. Technology & Architecture | Integration readiness, MLOps maturity |
| 4. Talent & Skills | Technical capability + org-wide AI literacy |
| 5. Governance & Ethics | AI policy, risk classification, accountability |
| 6. Research/Benchmarking Capability | Does the org track its own progress vs. peers? |
Five maturity levels
And the structural gap that shows up repeatedly in this region.
The weighted composite of the six dimensions maps onto a level. The level on its own is not the finding - the finding is the distance between where an organisation sits and where it is expected to operate.
The five levels, with the gap this region commonly carries
The dashed arrow is the gap: commonly sitting at level 1 to 2, expected to deliver at level 4. The plotted level is the composite, weighted across the six dimensions.
Government and semi-government entities here frequently carry a structural gap: mandated to deploy at level four while sitting at level one or two on data and talent. That is a compressed-adoption pattern, not a client failure, and naming it accurately builds more trust than either flattering or shaming the client. It also changes what the first project should be.
| Level | Label | Composite range | Typical profile |
|---|---|---|---|
| 1 | Unaware | 0–1 | No formal AI strategy; siloed data; ad hoc experimentation |
| 2 | Exploring | 1.1–2 | Isolated pilots, no real sponsorship, policy in draft |
| 3 | Developing | 2.1–3 | Some pilots live, sponsor named, governance forming |
| 4 | Scaling | 3.1–4 | Multiple use cases in production, governance enforced |
| 5 | Transformational | 4.1–5 | AI embedded in core processes, continuous benchmarking |
The four problem shapes
Used to select which proof is relevant to a prospect, and to decide what a first project should be.
Almost every documented engagement reduces to one of four. The value of the classification is not taxonomy for its own sake: it is that once a problem's shape is identified, the evidence for what works on that shape already exists, whatever industry it came from.
A human reads it and types it in again
Invoices, contracts, scanned paper, submittals
The answer exists and nobody can find it
Procedures, standards, specifications
Too much arrives to check it all, so you sample
Inspections, calls, feedback, camera hours
The same judgment, made differently every time
Risk scoring, prioritisation, approvals
The opportunity map
Impact against effort, with data readiness in the marker. The single most-referenced page when a report is circulated internally.
Two axes and a third variable in the marker fill, because data readiness is what actually determines whether a high-impact, low-effort opportunity is deliverable this quarter or next year. A hollow marker in the top-left quadrant is the most useful single thing an audit can produce: it says do this, and here is the reason you cannot yet.
- Estimated impact
- The vertical axis.
- Effort and complexity
- The horizontal axis.
- Data readiness
- Carried in the marker fill: solid is ready, part-filled needs work, hollow is not ready.
Start here
These are the instruments. The audit is the two to four weeks of using them on your operation.
Take the frameworks and run them yourself if that is more useful - they are published for that reason, and a client who self-assesses arrives at a better first conversation.
You get a reply within one working day, from the engineer who would do the work - not a sales sequence.