--- title: "30-day AI diagnostic for SMEs and mid-caps | GiLabs" description: "Your processes mapped, an impact × effort matrix and a costed 90-day roadmap. One fixed fee agreed before work starts, with a results guarantee." url: "https://gilabs.fr/en/services/ai-diagnostic/" lang: "en" service: "30-day AI diagnostic" --- # 30-day AI diagnostic: how much faster your processes can run *Service · entry point* Before you spend a euro on AI, you know what it will return. In 30 days: your processes mapped, the real end-to-end delay of each one, a return in hard euros per priority case and a payback measured in months. My method has three steps, map it, prioritise it, then automate it, in that order: automating a cluttered process does not speed it up, it multiplies it. That is why part of the gain I quantify for you cannot be bought, least of all from me. Quantified by the person who will build it. **Book a 30-minute call**: [Calendly](https://calendly.com/yann-gilabs/30min) ## This is for you if… - You run an SME, a mid-cap or a professional firm, and your teams lose hours to repetitive work: data entry, inbound documents, reporting, reconciliations. - Your day-to-day runs on Microsoft 365, SQL Server or line-of-business apps, and your data is usable. - You've been sold AI in promises, never in euros or payback time. - You want a documented investment decision, not yet another proof of concept that ends up in a drawer. ## What you receive - **Process map**: Where the hours go, on which documents, in which tools: in writing, validated with your teams. - **Impact × effort matrix**: Every use case positioned: expected gain, complexity, data prerequisites. Bad ideas are ruled out, in writing. - **The list of what you can stop doing**: The steps nobody in the company can justify any more, quantified. The one deliverable that pays you back without you having to buy anything. - **Costed 90-day roadmap**: The priority cases sequenced, each with its estimated return in euros and payback time. - **One brief per priority case**: Scope, technical feasibility on your IT system, prerequisites, cost estimate: enough to tender or build, with or without me. ## The process, week by week Three fixed milestones: kick-off, a mid-point review, and a final presentation at day 30. You always know where things stand. 1. **Map (weeks 1-2)**: Workshops with leadership, then with operational teams: we follow the hours, the documents and the tools, not the received wisdom. We also measure the time that actually elapses between the start and the end of a case, which rarely bears any relation to the time spent working on it. On site or remote. (Deliverable: process map, real end-to-end delay, and the full list of candidate use cases.) 2. **Prioritise and quantify (week 3)**: Every step has to justify its existence, or it becomes a candidate for removal: that is the gain you recover without buying anything. For what remains: the current cost of the process, the estimated cost after automation, and the implementation complexity on your actual IT system. (Deliverable: the list of steps to stop, impact × effort matrix, return in hard euros per case.) 3. **Decide (week 4)**: Final presentation to leadership: a sequenced 90-day roadmap, ruled-out cases with reasons, and a clear next step. You decide with full information. (Deliverable: costed roadmap + detailed briefs for the priority cases.) ## Investment A poorly scoped AI project burns tens of thousands of euros: pointless development, dormant licences, burnt-out teams. The diagnostic exists so that money is never spent in the wrong place. Investment: One fixed, all-in fee. Agreed in writing before work starts, with no commitment beyond the 30 days. Guarantee: if the diagnostic surfaces no opportunity with a payback under 12 months, I refund half the fee. ## Frequently asked questions ### We haven't done anything with AI yet. Is it too early? Quite the opposite: it's the best moment, and the most common situation. The diagnostic exists precisely to tell you where to start, and above all where not to start. It requires no AI prerequisites: only real processes and usable data. ### What happens after the 30 days? You own the roadmap, and it's actionable with any provider or in-house. If you continue with me, each case moves to fixed-fee development: scope, price and acceptance criteria agreed before work starts. ### We already have Copilot. Why a diagnostic? Copilot speeds up your teams inside Office; it doesn't redesign a process and doesn't reach into your line-of-business apps. The diagnostic quantifies exactly what Copilot doesn't cover, and tells you when Copilot is the right answer. ## On the same topic - [The PHP flaw had been known for two years. AI found the key in one summer.](https://gilabs.fr/en/blog/faille-php-ia/): Two critical PHP flaws, French government sites taken offline. The underlying defect had been known since 2024, and AI sped up its exploitation. The real risk: not knowing what runs on your servers. - [AI is not cutting jobs. It is closing the entry door.](https://gilabs.fr/en/blog/ia-porte-des-juniors/): Among 21-29 year-olds in France, permanent hires are down 28% in the occupations most exposed to AI. Total employment has not moved. What it means for an SME. - [My AI made up a price. My rule fits on one line.](https://gilabs.fr/en/blog/ia-prix-invente/): An AI read a free credit as a price, with no warning at all. What that near miss says about the framework a small business needs to use AI. ## After the diagnostic - [AI agents & business applications](https://gilabs.fr/en/services/ai-agents-automation/): The cases on your roadmap, built and shipped to production in your IT system at a fixed fee. - [.NET AI integration](https://gilabs.fr/en/services/dotnet-ai-integration/): For .NET estates: AI developed inside the application itself. **Book a 30-minute call**: [Calendly](https://calendly.com/yann-gilabs/30min)