Most companies are more ready for AI than they think in some areas and less ready in others. Readiness is rarely about having perfect data or advanced systems. It is about having a clear owner, simple rules, a real problem to solve and people prepared to change how they work. Score yourself honestly on the twelve points below: one point for each "yes".
Leadership and direction
- Is there one named person accountable for AI? Not a committee, not "IT", one person.
- Do senior leaders use AI tools themselves? Even occasionally, on their own work.
- Can you name the business problem you want AI to help with? In one sentence, with a rough number attached.
Rules and risk
- Is there an approved list of AI tools? Even if it has one item on it.
- Do staff know what data must never go into AI tools? Written in plain English, not buried in a policy.
- Is someone checking legal and data protection implications? For example, your data protection lead.
People and skills
- Has anyone had hands-on AI training on their real tasks? Not just an awareness webinar.
- Are there people in the business already curious about AI? Potential champions.
- Have you talked openly with staff about what AI means for their roles?
Process and data
- Can you identify a recurring, time-consuming task suitable for AI? Text-heavy, repetitive, rules-based work is a good sign.
- Is the information that task needs accessible? In documents or systems people can use.
- Could you measure that task today? Time taken, volume, error rate.
What your score means
| Score | Readiness | What to do next |
|---|---|---|
| 0 to 4 | Early | Appoint an owner, approve a tool, write data rules, train one manager |
| 5 to 8 | Developing | Pick one workflow, train the team, run a 90-day pilot with a baseline |
| 9 to 12 | Ready to scale | Standardise successful workflows, expand training, build a fuller strategy |
The most common gaps
In our experience, organisations most often fail points 1, 7 and 12: no single owner, no hands-on training and no way of measuring results. The good news is that all three can be fixed within weeks and without significant spending.
Points 11 and 6 are where technical or legal work may genuinely be needed, but for today's general AI assistants the requirements are usually lighter than people expect.
Readiness is not a gate
Do not treat readiness as something you must perfect before starting. A score of five is enough to begin a careful pilot, and running that pilot will raise your score faster than any planning exercise. Waiting for full readiness often means staff use unapproved tools in the meantime, which is the least ready position of all.
Our article on how to know if your business needs AI covers the "should we?" question; this checklist is about "are we able to?". If your gaps are mainly in leadership and skills, our Leading AI Innovation & Organisational Transformation course addresses points 1 to 9 directly and helps you plan the rest.
Frequently asked questions
Do we need clean data before using AI?
For general AI assistants used on everyday tasks, usually not. Clean, well-organised data matters much more for custom models and advanced automation.
Who should complete the readiness check?
Ideally a small group: a senior leader, a manager close to the work, someone from IT and someone from data protection. Different views give a more honest score.
How often should we reassess readiness?
Every six months, or whenever you plan a significant new AI initiative.