How do you build an AI strategy for a small business?

The three starting questions for an SME AI strategy

The first question — what problems are we trying to solve — is the most important and most skipped. Small businesses that start AI strategy by asking “which AI tools should we use?” are answering the wrong question. Start with: where does our business lose time, money, or quality? Where are our people doing work that does not require human judgment? Where are decisions being made with inadequate information? These are the candidate areas for AI application. The second question — where is AI genuinely better — requires honest assessment. AI is not always better. For some tasks it is dramatically better; for others it introduces risk without proportionate benefit. The third question — what do we need in place — prevents expensive mistakes: deploying AI before the governance, data, and skills foundations are ready creates tools that do not perform and risks that are not managed.

What to have in place before buying any AI tool

Before purchasing any AI tool, a small business needs four things in place. First, an AI use policy — even a simple one — that defines what is and is not permitted: what data can go into AI tools, who is authorised to use which tools, and what verification is required before AI outputs are acted upon. Second, an understanding of the EU AI Act obligations that apply to the specific tools being considered — particularly if those tools touch employment, credit, or client-facing decisions. Third, a named person responsible for AI governance — in a small business this is usually the managing director or a senior manager. Fourth, a clear definition of what success looks like for each AI application, so the business can evaluate whether the tool is delivering value. Building these foundations takes less time than recovering from a governance failure.