Key Takeaways
- Most AI projects fail for organisational reasons, not technical ones: MIT found 95% of enterprise GenAI pilots delivered no measurable profit impact.
- Readiness comes down to five things: a clear problem, decent data, documented processes, the right people and a sensible budget.
- AI is only as good as the data and process you give it, so tidy these first.
- Start small with a contained first project so you can prove value before committing further.
- Score yourself honestly. Fixing the gaps first is what makes a first AI project pay off.
Most AI projects that disappoint do not fail because of the technology. They fail because the business was not ready: messy data, unclear processes, or no one sure what problem they were solving.
This is not a hunch. MIT's 2025 research found that 95% of enterprise generative AI pilots delivered no measurable impact on profit, and the main cause was organisational, not technical: businesses could not fit AI into their workflows. This checklist helps you find out whether your business is ready before you spend a penny, so your first AI project lands on solid ground.
1. Do You Have a Clear Problem to Solve?
AI is a tool, not a goal. The readiest businesses can name a specific, costly problem, such as enquiries going unanswered or hours lost to admin. If your reason for adopting AI is that everyone else is, pause. Start from a problem worth solving, as we cover in how to build an AI strategy.
2. Is Your Data in Reasonable Order?
AI is only as good as the information you give it. You do not need perfect data, but you do need:
- Customer and enquiry records in one place, not scattered across inboxes.
- Reasonably accurate, up-to-date information.
- A clear idea of what you can and cannot share with a tool.
If your data lives in five disconnected systems, tidying that up is often the real first step.
3. Are Your Processes Documented?
You cannot automate what you cannot describe. If a task lives only in someone's head, write it down first. Clear, repeatable processes are the foundation that AI builds on. Automating a vague or broken process just creates faster chaos.
In our experience, this single step, writing down how a task is actually done, is where most of the value of an AI project is quietly created.
4. Do You Have the Skills and Buy-In?
AI changes how people work, so people decide whether it succeeds:
- Is someone responsible for making it work, not just buying it?
- Is the team open to it, or quietly worried about their jobs?
- Will you invest a little time in training?
Honest answers here predict success better than any feature list.
5. Have You Set a Sensible Budget?
Factor in more than the subscription: setup, integration, training, and time to adjust. Start small with a contained first project so you can prove value before committing further. Watch for the hidden costs of AI.
Your Readiness Scorecard
Count how many you can answer yes to:
- A clear, costly problem to solve.
- Reasonably organised data.
- Documented key processes.
- Someone owning the project and an open team.
- A realistic budget including setup and training.
Four or five: you are ready, start now. Two or three: fix the gaps first. Zero or one: do the groundwork before buying any tools.
Frequently Asked Questions
Does my small business need AI at all? Not for its own sake. If you have a repetitive, costly problem, AI can help a lot. If not, there is no rush.
What is the most common reason AI projects fail? Poor preparation: unclear goals, messy data, or undocumented processes. As MIT found, the technology is rarely the issue.
Where should I start if I am not ready? Tidy your data and document your most important process. That alone makes you far readier, and it has value with or without AI.
How much should I budget for a first AI project? Keep the first project small and contained, so the cost is modest and the value is easy to see. Factor in setup, training and a little time to adjust, not just the subscription.
How long does it take to get ready for AI? Often just a few weeks of tidying data and writing down key processes. That groundwork is the difference between a project that pays off and one that stalls.
The Bottom Line
Readiness for AI comes down to five things: a clear problem, decent data, documented processes, the right people, and a sensible budget. Score yourself honestly, fix the gaps first, and your first AI project is far more likely to pay off.
If you want an objective view of your readiness and a sensible first step, get in touch. We offer AI strategy consulting for UK businesses.
