- A readiness assessment scores five things: a real use case, your data, your processes, your people and your governance.
- It is worth doing because the barrier has moved off the technology: in a 2026 Publicis Sapient survey, 42% of large organisations said AI is capable but they are not set up to capture its value.
- A low score is a plan, not a verdict. It tells you exactly what to fix before you spend.
An AI readiness assessment is a scored check of whether your business can actually deliver an AI project, covering your data, your processes, your people, your governance and whether you have a real problem worth solving. It answers a plain question with evidence rather than optimism. The reason it has become the decisive step is that the technology stopped being the limiting factor. Among large organisations surveyed by Publicis Sapient in 2026, 42% said AI is capable but their organisation is not set up to capture its value. And in McKinsey's 2026 State of AI survey, 44% of respondents say AI is now scaling across their enterprise, up from 38% a year earlier, which leaves most still short of it. Readiness is what separates the firms that scale from the rest.
It sits alongside the AI audit, which finds the opportunities. The audit tells you what is worth doing; the readiness assessment tells you what you can do now and what needs work first. Run together, they turn "we should do something with AI" into a costed, sequenced plan.
The five things it checks
A real use case. Not "we want AI", but a specific, painful, repeated problem with a number attached. Vague ambition is the most common reason a project drifts and dies. If you cannot name the problem in a sentence, that is the first finding.
Data. The blocker named most often: 52% of respondents to PEX Network's 2025/26 industry survey of process professionals named data quality and availability as their biggest AI adoption challenge. The assessment checks whether the data your use case needs exists, is accessible, and is clean enough to trust. When it is not, the honest recommendation is to fix that before building anything on top of it.
Processes. Whether the work is understood well enough to hand part of it over. A process nobody has mapped cannot be automated safely, because you cannot tell the machine what "correct" looks like.
People. Whether someone can own the system once it is live. A common gap here: among EU businesses that considered AI but did not use it, 70.89% cited a lack of relevant expertise (Eurostat, 2025). The UK government's own AI adoption research names limited AI skills and expertise as one of the two reasons businesses most often give for not adopting it. You do not need a data science team, but you do need a plan for who runs and checks the thing.
Governance. Whether you know where a human stays in the loop and what happens when the system is wrong. This is what keeps a working pilot from becoming a quiet liability.
How to read a low score
A low readiness score is the most useful result you can get, because it is specific. It does not say "you are not ready for AI". It says your data on this particular use case needs cleaning, or your process needs mapping, or you need someone to own the output. Each of those is a task with a cost, not a closed door. The businesses that skip this step are the ones that fund a pilot, watch it stall, and end up in the MIT statistic: 95% of organisations getting zero return from generative AI. A weekend of honest scoring is a lot cheaper than that.
Small businesses have an advantage here
Readiness is not about size or budget. In many ways a smaller business is readier, because there is less legacy process and fewer competing systems to untangle. What it needs is a genuine problem, data it can reach, and a way to keep the result running. Our own companies are run on exactly that basis, and the plain version of what that looks like day to day is in a normal Tuesday. Size is not the test. Clarity is.
Common questions
What is an AI readiness assessment?
It is a scored check of whether your business can actually carry an AI project: your data, your processes, your people, your governance and a real use case. It answers a yes or no question with evidence, rather than assuming the answer is yes.
How is readiness different from an AI audit?
An audit finds the opportunities. A readiness assessment checks whether you can deliver them. In practice they run together: the audit says what is worth doing, the readiness check says what you can do now and what needs fixing first.
What is the most common reason a business is not ready?
Data. In PEX Network's industry survey, 52% of process professionals named data quality and availability as their biggest AI adoption challenge. The other common gap is a lack of in-house expertise to run and maintain what gets built, which Eurostat found was the leading reason EU businesses that considered AI did not adopt it.
Can a small business be AI ready?
Yes, and often faster than a large one, because there is less legacy to untangle. Readiness is about having a real problem, usable data and a way to keep the system running, not about size or a big budget.
If you would rather score your own business than read about scoring, the readiness checklist gives you the ten questions to run through, and the maturity assessment shows what the stages above "ready" look like. To have us run it with you, start at contact.
Sources
Every figure in this article links back to the source below it was checked against.
From the Orwell Lab archive, rebuilt in our current design. Figures checked against the sources listed above on 30 September 2026.
Discuss your own requirements ↗