What an AI readiness assessment is
An AI readiness assessment is a structured review of whether a business can adopt AI and get a return from it. Most versions cover five areas: data, technology, skills, governance and strategy. The result is usually a score or a maturity level, with a list of recommended use cases.
That format answers a narrow question: do you have the ingredients? It does not answer the question that decides whether the investment pays back: is the business organised to own, fund and govern AI once it is running?
What most assessments miss
AI projects rarely fail because the model did not work. They fail because nothing around the model changed. These are the structural questions a standard readiness score leaves out, and each one is a common cause of an AI investment that goes live and then quietly loses money.
Who owns it
Not a committee and not "IT". A named person accountable for AI decisions, standards and risk. Where accountability is spread across several teams, cost goes ungoverned and nobody notices when performance drifts.
Whether the commercial model can carry the cost
AI running costs rise with usage. A service priced on a fixed fee, with AI costs that grow every month, looks profitable in the first quarter and unprofitable by renewal. A readiness assessment should model cost at month one, month six and month twelve.
Whether the contracts cover it
Standard terms were not written for AI. Who is liable when the output is wrong, what you may do with client data, and what you must disclose all need an answer before a customer asks.
Who owns the customer once it is live
AI systems are not finished at go-live. They drift as real-world data changes and need monitoring and tuning. If nobody owns the account after the project team leaves, nobody sees the decline until the customer does.
What happens to the people
If AI changes what a role involves, the plan for reskilling and redeployment has to exist before the technology arrives. A business that finds this out afterwards loses the trust of the staff it needs to make the system work.
A business can score well on data, tooling and skills and still have no answer to any of these five. That business is not ready. It has the parts and nowhere to put them.
What the Keystone Diagnostic assesses
The Keystone Diagnostic is CXBridge's fixed-price AI readiness assessment. It looks at four things, in this order, because each depends on the one before it.
- Operating model. How authority, accountability and delivery are organised today, and whether that structure can absorb AI.
- AI readiness. Whether the governance, commercial, data and people foundations are in place to start.
- AI maturity. Where AI is already in use, how well it is being run and governed.
- Opportunities. Which use cases are worth pursuing, in what order, with the cost and return modelled on each one.
The assessment is led by Damian Copeland, an ISO 42001 Lead Implementer with 27 years of running customer experience and delivery operations. It is carried out with decision-makers directly. There are no junior consultants and no handover between a sales team and a delivery team.
What you get
One report, presented to your leadership team in person, not emailed as a document. It includes:
- A full assessment of the operating model, AI readiness and AI maturity
- A prioritised set of opportunities with cost, return and payback modelled for each
- An EU AI Act risk classification on every recommendation
- An indicative delivery programme
- An AI use policy written for your business
- An AI governance framework
Sometimes the conclusion is to slow down. If the foundations are not ready, the report says so, and sets out what has to be in place before money is spent on technology. That is a result, not a failed assessment.
Five questions to ask of any AI readiness assessment
These apply to any provider, including CXBridge. If an assessment cannot answer them, it is measuring ingredients.
- Does it name who is accountable for AI? A person, with the role written down.
- Does it model running cost over time? Usage-based costs at several points in the contract, not a single figure.
- Is the return on each use case modelled or asserted? Ask to see the working.
- Does it tell you what not to do? An assessment that recommends everything has prioritised nothing.
- Does it classify regulatory risk for each recommendation? Under the EU AI Act, some uses carry obligations that change the business case.
For a longer version to work through with your own team, download Before you invest in AI: ten honest questions.
Who it is for
UK SME and mid-market businesses, typically under 500 people, and business units inside larger organisations. Both are under pressure to invest in AI and want an honest view of whether the structure can carry it. It suits those that have not started, and those that have started and are not seeing the return.
Enterprise customers are not out of scope. What decides the approach is how fragmented the business is. A division with its own customers, delivery teams and P&L can be assessed as a unit in its own right. Where authority and accountability are spread across many units, the assessment is scoped unit by unit, because a single score for the whole group would hide the differences that matter.
It is not for businesses whose main aim is to cut staff through automation. CXBridge does not take that work. Where the assessment shows the operating model is the constraint, the next step is the design stage of the wider CX and AI engagement, built on CUSCOE.
Common questions
Is an AI readiness assessment the same as an AI maturity assessment?
No. Readiness asks whether the business can start and get a return. Maturity asks how well it runs the AI it already has. A business with no AI in production has nothing to score on maturity, so an assessment that reports only a maturity level is answering a question that business has not reached yet.
We already use AI tools. Do we still need an assessment?
Using tools and being organised to run them are different things. If nobody is named as accountable for AI, if usage costs are not tracked against margin, and if contracts do not mention AI, the business has adopted AI without becoming ready for it. Those gaps usually surface at renewal or at the first incident.
Does the assessment cover the EU AI Act?
Yes. Every recommendation in the Keystone report carries an EU AI Act risk classification. The Act can apply to UK businesses whose AI systems, or the outputs of those systems, are used in the EU, so the classification matters for most firms with European customers.
Will the assessment tell us to reduce headcount?
No. CXBridge does not take engagements designed to cut jobs through AI, and the assessment is not built to find them. Where AI changes a role, the report sets out reskilling and redeployment first.