Many companies want to "do something with AI" without first asking whether they're ready. Here are the 5 questions an executive should answer before signing any project.
1. Do we have data — and is it accessible?
AI runs on data. If your information is scattered across ten systems that don't talk to each other, in unsearchable PDFs or in spreadsheets a different person maintains every week, the first investment isn't AI — it's putting your data in order. The good news: you don't need a perfect data warehouse, but you do need a clear picture of what data you have, where it lives and how to access it.
2. Are our processes documented?
An AI agent learns from your documentation. If the procedure for handling a return only exists in the head of a senior employee with 12 years of tenure, the agent won't be able to learn it. Before automating, you have to document. The good news: documenting also benefits the company without AI — it reduces dependence on key people and accelerates onboarding.
3. Do we have minimum infrastructure?
You don't need supercomputers. But you do need: connectivity to your enterprise systems (APIs or databases), an environment where you can deploy services (cloud or on-prem), and a team that can do basic operations (deploys, monitoring, backups). If you're missing all three, the AI project will be slowed down by basic infrastructure issues.
4. Is there an internal champion?
AI projects fail when there's no executive owner. It's not enough that "everyone agrees" — there needs to be one person with authority, time and clear accountability. Without a champion, every decision becomes a meeting and every meeting becomes another meeting.
5. Are we culturally ready for change?
The hardest part of AI isn't technical — it's cultural. People need to learn to delegate to a non-human agent, accept that some processes will change shape and adapt their daily work. Companies with high turnover, hierarchical decision-making and resistance to change often have AI projects that fail not because the technology doesn't work, but because no one ends up using it.
What to do if you answered "no" to more than one
It doesn't mean you can't do AI — it means you need to prepare first. A solid 6-month preparation tends to save you 18 months of frustration later. If you want a structured assessment of your current readiness, our AI Strategy Consulting includes a 5-dimension diagnosis exactly like this one.
Want to learn more? Book your free session →