How do I know if my company is ready for AI?

You know your company is ready for AI when you can answer six questions honestly and not flinch at your own answers. Can you produce clean customer data on demand? Are your core processes documented well enough that a new starter could follow them? Would your team be capable of using an AI tool tomorrow? Do your systems talk to each other? Does anyone senior actually own the outcome? And could you explain your controls to somebody from outside the business? Readiness is those six things: Data, Process, People, Technology, Strategy, Governance. Nothing about model choice appears on that list.

The free AI readiness audit is 30 questions across those six pillars, five per pillar, and takes 7 minutes. At the end you get a score out of 120, a band for each pillar, and a short synopsis. That is the whole of the free tier. It is not an action plan, it does not inspect your systems, and every answer is self-reported by you. That last point is not a weakness to apologise for. The value of the exercise is that you are forced to state, in writing, things you have been carefully not stating in meetings.

Why self-reported answers still tell you something real

People assume a diagnostic is only worth having if somebody external verifies it. In practice the opposite problem is more common. Leaders know the answer and have never been made to say it. Ask a CFO whether the finance and CRM systems share data without a manual export and she will tell you the truth immediately, because she has watched someone do the export.

The self-report format also removes the thing that makes formal assessments slow. RSM sells its AI readiness assessment as a four-week engagement producing a roadmap and implementation plan. The four weeks are not spent discovering facts your own people do not already possess. They are spent gathering those facts from your people, reconciling them, and formatting them. If you want the roadmap, that engagement is a reasonable thing to buy. If you want to know where you stand before you spend anything, the diagnosis is thirty questions you can answer yourself.

What a score out of 120 buys you is a shape. Six bands next to each other tell you which pillar is dragging. Most businesses find one pillar sitting well below the rest, and it is almost never Technology.

Data: the 500-customer question

The audit’s first question is this: if you needed a clean, accurate list of your best 500 customers right now, how confident are you in what you would get?

Run it as a live test rather than a thought experiment. Ask for the list by end of day. Time how long it takes and count how many people touch it. Every hand it passes through is a place where a definition gets applied differently, and every judgement call about what “best” means is a judgement an AI system will make for you, invisibly, and at scale.

The other four Data questions push on completeness, whether key systems can share data without someone manually exporting and importing spreadsheets, how often reliable data actually drives an important decision rather than being retrofitted to one, and who is responsible for data quality by name. If you cannot answer that last one, the cheapest AI mistake a business makes is already in progress and has nothing to do with AI.

Process: ask three people to do the same task

The diagnostic question here is deceptively mild. If you asked three different people in your business to complete the same task, how similar would their approach be?

Try it this week. Pick something routine, give it to three people separately, and compare what comes back. Divergence is not a sign of a bad team. It is a sign the process lives in three heads rather than one document. Automation applied to an undocumented process does not tidy it up. It picks one of the three versions, usually the one the implementation partner happened to observe, and runs it a thousand times a day.

The pillar also asks how many core processes are documented well enough that a new team member could follow them without asking anyone, how often you review and improve processes rather than just running them, how clearly you can name where the bottlenecks sit, and who is responsible for making sure processes are followed and improved over time. Five questions, no technology in any of them.

People: what your team actually thinks

The first People question asks you to describe your team’s attitude toward AI honestly. Most leaders answer optimistically and are wrong. The second asks whether, if you told your team to use an AI tool in their work tomorrow, they would be capable of it.

The third is the uncomfortable one. As the leader, how actively are you modelling AI adoption: trying it yourself, talking about it, normalising it? A leadership team that has delegated AI entirely to IT has already told the organisation what it thinks the priority is. The pillar closes on whether the business has made any formal investment in helping the team understand and use these tools, which for many companies is a one-word answer.

This pillar is where businesses without a tech team often score better than they expect. Capability and curiosity are not the same as headcount.

Technology: what the systems can actually do

This is the pillar people assume the whole audit is about, and it is one sixth of it. The questions cover the state of your core systems, integration, and whether the infrastructure you have could support something new without a rebuild.

Be aware of what the audit does not do here. It does not inspect your stack. It does not review the tools you have already paid for or check whether anyone still uses them. Your answers are your answers. If you overstate your integration maturity you will get a flattering band and a useless one.

Strategy: what the AI is for

Strategy asks whether AI work connects to a business outcome somebody is already accountable for, or whether it exists because the board asked about it. This is where the pilot problem originates.

The numbers are unambiguous. PwC’s 29th Global CEO Survey, covering 4,454 leaders across 95 countries, found 56% of CEOs reported neither higher revenues nor lower costs from AI over the previous 12 months, with only 12% achieving both. A survey of 650 enterprise technology leaders found 78% have AI agent pilots and 14% reach production scale, and RAND put the share of enterprise AI projects failing to deliver promised business value at 80.3%.

A successful pilot proves the model works on a problem somebody chose because it was easy to demonstrate. It proves almost nothing about whether the organisation can run the thing at scale. That is why pilots stall on the way to production rather than at the proof-of-concept stage.

There is a timing problem sitting underneath this too. Open Future Forum found 72% of CEO and founder respondents expect AI payback inside six months against 45% of finance-seat respondents, a 27-point gap between the seat that signs and the seat that proves. If your board and your CFO are arguing, they may not be arguing about AI at all. They are arguing about the clock.

Governance: say the name out loud

Here is the test. Say, out loud, the name of the one person accountable for AI outcomes in your business. If it takes you more than three seconds, or if the name changes depending on which failure you imagine, you do not have an owner. You have a committee that meets.

This is the most common gap in the market. VentureBeat survey data found 85% of enterprises run contested AI platforms with no single owner, and only 10% can automatically detect a failing model. When those leaders were asked for the biggest obstacle to governing AI across platforms, “no single owner or accountable team” came second at 29%.

Don’t expect the board to close the gap for you. Diligent’s 2026 research found only 8% of directors report strong AI expertise on their board, the lowest of every area surveyed, while 44% want AI risk and opportunity raised to the top of the agenda. BCG’s survey of 625 CEOs and board members found 60% of CEOs think their boards are too impatient with the pace of AI change. A board that wants AI at the top of the agenda and cannot assess it is not impatient. It is uninformed, and it will keep asking for pace because pace is the only variable it can see.

The question of who owns AI outcomes in a company is separable from whether you need a chief AI officer. Most businesses under 10,000 people do not. They need one existing executive whose objectives include the outcome, not the technology.

What to do with the score

Six bands, one per pillar. Look for the lowest. That is where the next quarter’s work is, and it will usually be Data or Process, which is inconvenient because neither photographs well in a board pack.

If you want the fixes rather than the diagnosis, the Full Report is £97 and includes action plans. The Deep Dive at £497 is a 30/60/90-day plan for closing the gaps before you spend on AI itself. Both sit on the same diagnosis. The free score tells you your position. The paid tiers tell you the route.

One warning on self-assessment. The score is only as honest as the answers. Take it alone, answer as though nobody will see it, and resist the instinct to round up.