What does an AI readiness audit measure?

An AI readiness audit measures the business conditions that determine whether AI will work, rather than the AI itself. The AI Readiness Partner audit measures six pillars with five questions each: Data (is your information accurate, complete and shareable between systems), Process (are core processes documented and consistent), People (is the team willing and capable), Technology (can your estate support something new), Strategy (is AI activity tied to an outcome and an owner) and Governance (do rules and accountability exist). Thirty questions in total, scored out of 120, with a band for each pillar so you can see which dimension is holding you back rather than receiving a single ready-or-not verdict.

How long does a free AI readiness audit take and what do you get?

The free AI Readiness Partner audit is thirty questions and takes 7 minutes. At the end you get a readiness score out of 120, a band for each of the six pillars (Data, Process, People, Technology, Strategy, Governance) and a short synopsis. That is the complete free output. It does not include an action plan, and no sales call is booked unless you ask for one. The paid tiers sit on the same diagnosis: a £97 Full Report adds action plans, and a £497 Deep Dive is a 30/60/90-day plan for closing the gaps before money goes on AI tooling.

Does an AI readiness audit inspect our systems or data?

No. A readiness audit of this kind is a structured self-assessment, not a technical inspection. Nobody connects to your CRM, samples your database or reviews your code. Every answer is self-reported by the person taking the audit, which means the accuracy of the score depends entirely on how honestly it is answered. It also does not review tools you have already bought or whether anyone still uses them. What it does is arrange your own read of the business into six comparable dimensions with a band each, so the weakest area is visible and arguable rather than merely suspected.

Is an AI readiness assessment useful for a mid-sized company rather than a large enterprise?

Yes, and arguably more so. Most AI readiness frameworks are built for enterprises with dedicated data teams and a chief data officer, so they measure things a 200-person business does not have. A six-pillar assessment sized for businesses between roughly 50 and 10,000 employees asks answerable questions instead: whether three people in the same team would complete a core task the same way, whether systems exchange data without manual spreadsheet exports, and who is named as responsible for data quality. Those questions expose the gaps that actually stop mid-sized AI projects, and they can be answered by one leader in 7 minutes without commissioning a consultancy.

7 September 2026

What does an AI readiness audit actually measure?

A plain explanation of what an AI readiness audit measures: six pillars, thirty questions, a score out of 120, and what it deliberately does not do.

By Carl Chessum

An AI readiness audit measures six things about your business, and none of them are AI. It measures whether your data is trustworthy, whether your processes are repeatable, whether your people are willing and able, whether your technology can pass information between systems, whether your strategy names an owner and an outcome, and whether anyone has written down the rules. Six pillars: Data, Process, People, Technology, Strategy, Governance. Five questions each. Thirty questions in total, scored out of 120, with a band for each pillar so you can see which one is dragging.

That is the whole mechanism. It is a structured self-assessment, not an inspection. Nobody connects to your CRM, nobody reads your code, nobody logs into your data warehouse. Every answer is given by the person taking it, which means the audit measures your own honest read of your organisation across six dimensions and then arranges that read into a shape you can take to a board meeting. The free version takes 7 minutes and gives you a score, a band per pillar and a short synopsis. It does not give you an action plan. I will come back to that, because the limits matter as much as the contents.

Why six pillars and not one number

Most AI readiness content collapses into a single verdict. Ready, or not ready. That is useless, because businesses are never uniformly one or the other. I have worked with companies with immaculate financial data and no documented process anywhere in the building. I have worked with companies where the processes were written up beautifully in a wiki nobody had opened since 2019.

A single score hides that. Six banded scores do not. If your Data band comes back weak and your People band comes back strong, you have a different problem, and a cheaper one, than a company with the reverse. Gartner expects organisations to abandon 60% of AI projects that lack AI-ready data through 2026, and in Informatica’s CDO Insights 2025 survey of 600 global data leaders, 43% named data quality, completeness and readiness as the top obstacle stopping AI initiatives reaching production (Gartner). If that is your weak pillar, the fix is unglamorous and known. If your weak pillar is Governance, the fix is a document and a name, and it costs a morning.

The pillar structure is also what makes the audit legible to a CFO. “We scored 71 out of 120, with Data and Governance in the bottom band” is a sentence a finance director can act on. “We’re not quite AI ready yet” is not.

Pillar one: Data

The Data questions test whether the information your business runs on could survive contact with a model. The first one is the one that gets the strongest reaction in a room:

If you needed a clean accurate list of your best 500 customers right now, how confident are you in what you would get?

People laugh at that question, then stop laughing. It is a proxy for about nine other things: whether your definition of “best customer” is agreed, whether the CRM is current, whether anyone maintains it, whether the answer arrives in an hour or a fortnight.

The other four in this pillar cover completeness (“How complete is the data your business relies on to make decisions?”), integration (“Can your key business systems share data with each other without someone manually exporting and importing spreadsheets?”), whether reliable data actually drives important decisions or merely accompanies them, and ownership: “Who in your business is responsible for the quality and accuracy of your data?”

That last question sorts businesses fast. If the honest answer is “nobody, really”, you have found something more important than a technology gap. I have written about why dirty data is the cheapest AI mistake to fix first, and the reason it is cheapest is that it needs no vendor.

Pillar two: Process

AI automates process. If the process is undefined, you are not automating it, you are freezing one person’s habits into software. The Process pillar measures whether there is anything solid to automate.

How many of your core business processes are documented well enough that a new team member could follow them without asking anyone?

If you asked three different people in your business to complete the same task, how similar would their approach be?

That second one is worth running as a live test before you answer it. Pick a core task. Ask three people in the same team to talk you through their steps. If the three accounts diverge in more than a trivial way, the process does not exist as a process. It exists as three habits with a shared name.

The remaining three questions ask how often the business reviews and improves core processes rather than just running them, how clearly you can point at the bottlenecks in your most important processes, and again the ownership question: who is responsible for making sure core processes are followed and improved over time.

Pillar three: People

This pillar is about willingness and capability, in that order, and it includes a question about you.

How would you honestly describe your team’s attitude toward AI right now?

If you asked your team to use an AI tool in their day-to-day work tomorrow, how capable do you think they would be?

As the business owner or leader, how actively are you modelling AI adoption, trying it yourself, talking about it and normalising it for your team?

That third one is uncomfortable by design. Leaders who have never opened the tool they are asking three hundred people to adopt tend to score their own organisation optimistically. The pillar also asks whether the business has made any formal investment in helping the team understand AI, which is a different question from whether anyone has sent round a link to a webinar.

Capability without willingness stalls. Willingness without capability produces enthusiastic mess. The audit measures both separately so you can see which you have.

Pillars four, five and six: Technology, Strategy, Governance

Technology is not asking what you have bought. It is asking whether your existing estate can move information around, whether anyone can support it, and whether you have the internal capacity to run something new. If you have no in-house technical people at all, that is a constraint rather than a disqualification, and I have set out what a business with no tech team can do first elsewhere.

Strategy asks whether AI activity is attached to a business outcome and a named owner. This is where the honest scoring hurts most, and there is external evidence for why. In August 2026 data from the Open Future Forum, business-unit sign-off on AI purchases fell from 18% to 8% inside a single month while “no single owner yet” doubled to 14%, and 29% of technology respondents reported an unowned AI purchase at their company against zero CEO or founder respondents (Open Future Forum). Read that gap carefully. The people at the top did not know about the purchases the people one level down were describing. If you want the longer argument, see who owns AI outcomes in a company.

Governance asks whether the rules exist. Not whether they are sophisticated. Whether they exist, whether anyone knows them, and whether anyone is accountable when a model produces something wrong or a staff member pastes client data into a public tool. Most mid-sized businesses score badly here and are surprised, because governance feels like an enterprise concern until the first incident.

What the audit does not measure

This is the part vendor pages leave out, so I will be blunt about it.

It does not inspect anything. No system access, no data sampling, no code review. Every one of the thirty answers is self-reported by the person sitting at the keyboard. If you answer generously, you get a generous score, and the score will be wrong in a way that costs you money later. The audit is only as good as your willingness to mark your own homework honestly.

It does not review what you have already bought. It will not tell you that the licences you renewed in March are being used by four people out of ninety. That question matters, and CIO reporting suggests it matters widely: three-quarters of CIOs surveyed said they had remorse over at least one major AI vendor or platform selection made in the past 18 months (CIO.com). But a readiness audit is a diagnosis of the conditions, not an inventory of the cupboard.

It is not a call, a consultation or a demo. There is no sales call at the end unless you ask for one.

And at the free tier it does not produce an action plan. You get a readiness score out of 120, a band per pillar and a short synopsis. That is genuinely all. If you want the action plans, the £97 Full Report has them, and the £497 Deep Dive is a 30/60/90-day plan for closing the gaps before you spend on AI. Same diagnosis underneath. The paid tiers are the course, not the position fix.

Why a self-reported score is still worth 7 minutes

Because the alternative most businesses use is no measurement at all, and the failure rate of that approach is documented. RAND research published in 2025 found 80.3% of enterprise AI projects fail to deliver promised business value: 33.8% abandoned before production, 28.4% reaching production but missing expected value, 18.1% running without ever recovering the investment (RAND, 2025).

Meanwhile the tolerance for finding out slowly has evaporated. In CloudZero’s survey of 260 finance executives including 135 CFOs, 87% said they need to tie AI spend to business outcomes within the year, and only 22% can do it today (CloudZero). Teneo’s May 2026 research found 65% of CEOs report misalignment with their CFO on AI’s long-term value (Teneo via wndyr).

A thirty-question self-assessment does not resolve that argument. What it does is give both sides the same six numbers to argue about, before the invoice rather than after. Answered honestly, it will usually tell you something you already suspected and had not written down, which is exactly the point: suspicion is not evidence, and a board cannot act on suspicion.

If you want your own six bands, the free 7-minute AI readiness audit is thirty questions and a score out of 120. Answer it as if the CFO is reading over your shoulder. That is the version worth having.