An AI engagement in an SME fits in five stages, and you decide between each.

Our engagement method follows five stages: Diagnostic, Scoping & compliance, Pilot, Deployment, Managed operations. Between each, a decision point: the next stage is only committed on the proof produced by the previous one.

You decide at each step, on figures measured at your premises, and one of the possible outcomes is always to stop: the deliverables of each stage are yours to keep.

The rule that runs through all five stages is the rule of our whole practice: the machine reads, reconciles and alerts; decisions stay human.

The diagnostic maps what actually happens in the company: what your teams use, the AI features active in your tools, the data that leaves, and the hours that searching for information costs.

It lasts four weeks and leaves five deliverables: a usage register, a data flow inventory, a search cost measurement, a drafted AI usage policy, a costed action plan. Everything belongs to you, with or without a follow-up.

Scoping turns findings into rules: company accounts configured, policy signed, teams trained, register maintained.

Compliance is handled from the design stage: the documented training covers the AI literacy measures required since February 2025, and the register prepares the EU AI Act obligations that really concern you.

The pilot installs the engine on a scope chosen to be demonstrable: restricted, reversible, measurable. It is demonstrated on public data before it touches yours.

The decision to extend is taken on measured results at your premises, never on a promise. If the pilot does not prove, we stop there, and the scoping stays with you.

Deployment connects the proven engine to your real flows and your teams: thirty seconds for what your teams currently spend thirty minutes on.

The automations that pay off are connected one by one, priced per unit: pre-sales, tenders, document production, reconciliation of management documents.

An AI system is not a frozen piece of software: models evolve, your data changes, an unmaintained infrastructure degrades. Managed operations supervise, update and re-index, under a written service commitment.

Every lesson learned feeds the action plan: what has proven extends, what has not proven stops. And at every stage, the same neutrality: we are paid for the project, never for reselling licences.