Are You AI Ready? The 20 Questions Every UK Business Leader Must Answer Before Investing

When an AI project fails — and the majority of first attempts do — the post-mortem almost always points to the same things. Not the technology. Not the vendor. The foundations.

Scattered data. Undocumented processes. Staff who were not involved in the decision. Leadership who signed off without understanding what they were buying. No definition of what success looked like before a single pound was spent.

AI automates processes. If you cannot articulate your process, AI will automate your chaos. That is not a metaphor. It is a description of what actually happens.

The 20 questions below have been developed from direct experience working with UK SMEs and councils. They are divided into four sections. Answer them honestly. The score at the end will tell you more than any vendor demo.

Can you find a specific customer record in under 60 seconds?

Not usually. Not if Sarah is in the office. Right now, from any device. If the answer is no, your data is too disorganised for AI to use reliably. Fix the data before you buy anything else.

Do you know where all your data actually lives?

List every system. CRM, accounting software, shared drives, individual laptops, cloud storage, paper files. If you cannot list them, AI cannot access them properly. Start with an audit.

Have you tested a data backup in the last six months?

AI projects introduce new points of failure. If you cannot restore your systems from a backup, you are not ready for more complexity. Test the backup. Today, if necessary.

Are your top three business processes documented step by step?

Sales process. Customer onboarding. Order fulfilment. Written down, not just understood by the people who do them. If the process lives in someone's head, automation is impossible.

Are your systems cloud-based or accessible via API?

AI needs to communicate with your systems. If your critical data lives on a server from 2012 in the back office, you are looking at infrastructure work before AI can help. That work is worth doing. But plan for it.

Has your leadership team had a structured AI discussion in the last three months?

Not a passing mention of ChatGPT. A proper conversation about what AI means for your business, your people, and your customers. If leadership is not engaged, AI becomes an IT project. IT projects fail.

Do you know whether your staff are already using AI tools?

They almost certainly are. ChatGPT, Copilot, Grammarly. Whether you know about it or not. If you do not have a policy, you have a risk. That risk is manageable. But only if you acknowledge it.

Can you name one internal person who would champion AI adoption?

Someone curious, competent, and respected by colleagues. Not the IT manager. Someone who understands the business and can translate AI into practical change. If you cannot name them, your adoption will stall.

Have you addressed staff concerns about AI affecting their roles?

Fear of redundancy will sabotage any AI project if it is left unaddressed. This does not require reassurance. It requires honesty and involvement. Staff who help identify AI use cases are far more likely to support the outcome.

Do you have a formal annual training budget and plan?

Organisations that do not invest in training now will not invest in AI training later. If your staff development is ad hoc, your AI capability will be too.

Can you articulate your business strategy in a single sentence?

"We help X achieve Y better than Z because of W." If you cannot say it simply and specifically, AI will not create the clarity that is missing. It will accelerate confusion.

Do you have a prioritised list of your top five business problems?

Not annoyances. Problems that are costing you money, customers, or sleep. AI solves specific problems. If you do not know which problems matter most, you cannot choose the right tools.

Have you reviewed a technology vendor contract in the last 12 months?

AI vendor contracts are written by lawyers working for the vendor. They lock in pricing, claim rights over your data, and make exit expensive. If you do not read contracts carefully now, you will be trapped later.

Does your leadership team understand GDPR Article 22?

It governs automated decision-making. If AI makes decisions that affect people — hiring, credit, access to services — you are legally required to understand this. Ignorance is not a defence. It is a liability.

Do you have budget tolerance for a first AI project that may require iteration?

First attempts often fail, or fail partially, or need reworking. If every pound of AI investment must deliver immediate ROI, you will either choose the wrong project or declare failure prematurely. Build in tolerance before you start.

Do you know your current cost per transaction, lead, or customer?

Without baseline metrics, you will never know whether AI actually saved you money. "It feels faster" is not a business case. Establish the baseline first, then measure against it.

Have you quantified the cost of your three most time-consuming manual processes?

Manual data entry. Chasing approvals. Formatting reports. Put a number on the hours and multiply by salary cost. That calculation is your AI business case. It is usually compelling.

Does your cyber insurance cover AI-related incidents?

Most policies written before 2023 do not cover prompt injection, AI data leakage, or hallucinations used in decisions. Check the policy. Update it if necessary. AI introduces new risk categories that did not exist when your current policy was written.

Could your business operate for 90 days if your primary AI-powered system went down?

AI dependencies create single points of failure. If your invoicing, customer service, or operations rely on an AI system, what happens when it breaks? Have a fallback. Know how long you can sustain it.

Have you defined what success looks like for your first AI project?

Specific. Measurable. "Save five hours per week on invoice processing" not "be more efficient." If you cannot define success before you start, you will never know whether you achieved it. And you will not be able to justify the next investment.

17 to 20 Yes: You are highly prepared. Start with a small, contained pilot project. Clear metrics, low risk, high impact. There are opportunities waiting.

12 to 16 Yes: You have a solid foundation. Address the specific gaps before committing to major investment. The work required is targeted and achievable.

7 to 11 Yes: Work is needed before AI tools are purchased. Focus on data, process documentation, and governance. This is not failure. Most organisations I work with start here.

0 to 6 Yes: Do not buy AI yet. You will waste money. The good news is the path to readiness is clear and the foundational work benefits your business with or without AI.

If you scored above 12, book a call. We will identify the right first project together, define success criteria, and build a 30-day plan.

If you scored below 12, we can help you build the foundations. The organisations that invest in readiness first get far better results from AI than those that rush in. That is not my opinion. It is the pattern that repeats every time.

No pitch. No obligation. Thirty minutes of practical advice.