Where to start with AI in an SME
Most SMEs start from the tool ("let's get ChatGPT for everyone") instead of the problem. The result is occasional use and no measurable impact. The right path is the opposite: first find the activities that waste time, then choose the tool.
What you can do, in practice
Map the repetitive activities
Who copies data from one system to another? Who prepares the same report every Monday? Who always answers the same questions? That is where the first return is.
Check where the data lives
ERP, Excel, email, shared folders: AI only works well if the data is reachable. Often the first step is simply to extract it and put it in order.
Choose 2-3 use cases
Simple criteria: impact (hours or euros saved), feasibility (data available), risk (what happens if the AI gets it wrong).
Run a pilot, not a project
4-8 weeks, one department, one measurable goal. If it works, you extend it; if not, you have spent little.
Set minimum rules
Which data can be given to the AI, which tools are authorised, who checks the output. One page is enough to start.
Where to start
A half-day workshop with the department heads to list the most repetitive activities, followed by a quick feasibility assessment. The output: a ranked list of use cases with estimated costs and benefits.
How I can help
- AI assessment: I interview the departments, look at the systems and the data, and hand you a concrete, prioritised roadmap.
- Turnkey pilot: I build the first use case together with your team, with measurable goals.
- Company AI policy: clear, workable rules on tools, data and responsibilities.
- Targeted training for the people who will actually use the tools.
Want to find out what can be done in your company?
A first half-hour chat, no strings attached: you tell me how you work and I'll tell you where AI can deliver a concrete result, and where it isn't worth it.