Part of our guide: AI strategy
An AI readiness assessment sounds straightforward, and the question it appears to answer is whether your business is ready for AI. That question is almost meaningless on its own, because it never establishes what you would be ready for, or what you are being compared against.
Most consultancies use maturity models, asking how mature your data capability is, how mature your technology is and how mature your leadership is. They score you on a scale and hand you a report saying you're at level 2.5 out of 5, leaving you with no sense of what to do next.
That approach is tidy and auditable, and it is almost useless for decision-making, telling you nothing about where to invest or what your business could actually do with AI.
A real AI readiness assessment maps your business capabilities against AI opportunity, telling you where AI creates real leverage for your business model and what is actually blocking you.
What a generic maturity model does:
Most maturity models ask questions like these: whether you have centralised data, a data governance framework, senior sponsorship and AI skills.
You answer those questions candidly, you get scored and you end up with a report that probably says: "You need better data architecture, clearer governance and more technical talent."
That report does not tell you whether better data architecture actually matters for your business model, whether governance will help you or slow you down and whether those technical people will actually move the needle.
For many mid-market businesses, the constraint is clarity on what problem AI actually solves rather than data or governance, and you could have perfect data and still be solving the wrong problem.
What a real assessment actually involves:
Capability mapping. This step establishes what your business is actually good at and where you hold competitive advantage, mapping what you actually do and what drives your business model. For insurance, that might be "claims handling" or "customer underwriting", and for financial services it might be "regulatory compliance" or "fraud detection".
AI opportunity mapping. For each capability, you look at where AI could create leverage: reducing cost, improving customer experience, speeding up decision-making or opening a new revenue stream. You are not looking for generic opportunities here, and what counts is leverage that is specific to your business.
Capability scoring. For each opportunity, you work out what is actually blocking you: data quality, technical skills, process complexity, regulatory uncertainty or senior leadership disagreement, and you score those blockers candidly.
That gives you a picture that reads "We could transform claims handling with AI, but we'd need better data integration and clearer regulatory interpretation". It might equally read "We could automate 60% of our compliance reporting, but we need someone who understands both the processes and the technology".
Strategic clarity. From that assessment, you identify the opportunities to tackle first, choosing the ones where you can actually execute and where the payoff matters rather than the easiest ones.
Why this matters differently for mid-market
A large insurance company with 500 people in technology might do "improve claims handling" and "improve underwriting" and "build a new distribution channel" and "optimise pricing" all simultaneously, and they have both the people and the infrastructure.
A mid-market business with 100-150 people in technology cannot do that, and you have to sequence ruthlessly, so an assessment that tells you "you're at level 3 maturity" does not help you sequence. An assessment that says "you can transform claims with 6 people and a six-month programme, which will reduce cost by 15% and improve handoff time by 40%" actually helps you decide.
Red flags in assessment approaches:
The pitch that says "We'll assess you against industry best practice" never establishes whose industry it means, or whose "best practice" it holds you against. If your competitor is three years ahead of you on technology, copying them gets you to where they were, not where you need to be.
Another pitch runs like this: "Here's our maturity model. Everyone fills it out." Generic models of that kind are scalable rather than thoughtful, and for a 200-person business a generic maturity model is expensive waste.
A third pitch promises "We'll interview 50 people across your business", and interviewing is fine in itself. If the assessment amounts to a record of what people said without connecting those insights back to business strategy and opportunity, you have paid for a survey, not an assessment.
A fourth pitch says "Your assessment will take 12 weeks", and it should not, because a real assessment of a 200-person business should take 2-3 weeks at most. If it takes longer, you are over-investigating, not getting clearer.
What a good assessment looks like as an output:
A good assessment hands you a clear, ranked list of AI opportunities. The first entry reads "First priority: automate X because it affects Y customers and costs Z annually", and the second reads "Second priority: improve X because it supports our strategic shift to Y".
It gives you a clear-eyed assessment of readiness for each opportunity, because you are not "ready" or "not ready" in the abstract. You are ready for this opportunity with these people and these capabilities.
It gives you a sequenced roadmap that reads "Do opportunity A first. It builds foundations for opportunities B and C. Opportunity D is independent, but harder, so do it later."
It sets clear success metrics that go beyond "improve efficiency", naming targets like "Reduce process time by 40% in quarter two, generate £X savings by quarter four".
It sets out realistic resource requirements that go further than "you need better data and more skills in the abstract". The requirement reads "You need one engineer for six months, a process redesigner for three months and external validation from X for £50k".
This is what our Breathe engagement does
Our five to ten working day discovery sprint maps your business capabilities, identifies genuine AI opportunity, assesses what is actually blocking you and hands you a prioritised roadmap. It is a strategic clarity sprint focused on what you can actually do, not a 12-week maturity assessment.
You come out of it knowing what to build, in what order, with what resources and why it matters.
Get in touch if you want an assessment that actually points toward action.