M
Meaning
What real problem are we solving?
MAKIA LABS · CUSTOM AI TRAININGS for teams
Generative AI. For real this time.
We train your teams to use ChatGPT, Claude, and Gemini on their real work — not demos. The MAKIA method. 60% hands-on. Measured outcomes.
Deployed at 30 industrial SMEs · 2025–2026
−30%
Operational costs
+15%
Revenue activities
+5h
Saved per manager / week
Averages observed by Makia Labs across 30 industrial SMEs (2025–2026). Varies with adoption depth.
OUR METHOD
Five steps. None of them skipped.
From mapping to adoption, a linear journey built for operational teams.
- 01MMap
Map real usage and friction points.
- 02AActivate
Spark the first high-value use cases.
- 03KKickstart
Quickly prototype: Project, GPT, Gem.
- 04IIterate
Learn from the field, adjust, measure.
- 05AAdopt
Embed into practice. Internal usage charter.
OUR COMPASS
Five questions before you deploy.
We never train on a tool without answering these five questions first.
A
Actors
Who will use it, and who decides?
K
Knowledge
What data, what expertise?
I
Impact
How do we measure the value created?
A
Alignment
Aligned with your culture & GDPR?
This filter eliminates 80% of 'AI-for-AI's-sake' projects. We keep the 20% that actually create value.
TAKE THE TEST
Want to see where your team stands on these 5 dimensions?
The MAKIA audit: 25 questions, 5 minutes, a personalised report by email. Free, no strings attached.
GDPR-compliant · Data stored in the EU
SERVICES
Four ways to move forward.
01
AI Readiness Audit
2 weeks. Interviews with your teams, mapping of priority use cases, AI maturity scoring against the MAKIA framework, audit of available data and GDPR constraints, then delivery of a 30/60/90-day roadmap with estimated ROI and a sequenced deployment plan.
Explore this service02
Custom in-house trainings
1 to 5-day training programs tailored to AI literacy level: fundamentals of generative AI, prompt engineering, main general tools (ChatGPT, Claude, Gemini), then specific tools and use cases matched to each team's roles and tasks.
Explore this service03
Prototype sprints
1–2 weeks to go from idea to a working Project / GPT / Gem: needs scoping, system prompt design, integration of your data and reference documents, iteration with end users, then handover of a version tested in real conditions with usage documentation.
Explore this service04
Executive briefings
2h with your exec team or board. State of the art of generative models, what really changes in your sector, what's just hype, framing of risks (GDPR, sovereignty, vendor lock-in), and concrete recommendations on 2026 AI budget allocation and the governance to put in place.
Explore this serviceIdéal pour : PME industrielles de 20 à 250 personnes dont les équipes paient déjà des outils IA mais les sous-utilisent.
TOOLS WE COVER
- Claude
- ChatGPT
- Gemini
- NotebookLM
- Notion AI
- Make
- Zapier
- n8n
- Lovable
- Dust
Most importantly, we teach you how to choose between them.
TRUSTED BY
Industrial SMEs, Innovative Startups, Public Sector.
CASE STUDY · 01 / 04
Cayrol Energie — Energy · Photovoltaic & hydroelectric · Frontignan
1-day training, May 12, 2026. Leadership & engineering. MAKIA framework applied to meeting notes, client emails, and tenders.
EnergyIndustryPublic sectorEducationAgricultural cooperatives
ABOUT
One standard: it has to work on Monday morning.
Makia Labs started from one observation: most corporate AI trainings stop at the demo. Teams leave impressed and… change nothing. Our MAKIA framework and CRAFT prompt method (Context · Role · Action · Format · Tone) are built for the opposite: you leave the room with working prompts, a configured Project, and a 90-day plan.
Research-grounded, active pedagogy, and one stated bias: we'd rather find one use case that saves 30 minutes a day than show ten slides about the singularity.
FOUNDER
Maria Angel Ferrero
PhD Innovation Management · Design Thinking Expert · AI Educator
10+ years of research in innovation, creativity, and digital transition. Built the MAKIA framework and AI fluency programs for students, startups, and executive teams.
IA EN PME
L'IA générative en PME : par où commencer, sans y perdre six mois
On accompagne des PME industrielles, énergétiques, agricoles et publiques qui veulent passer de « on a testé ChatGPT » à des usages réels, mesurés, conformes RGPD. Pas de promesse de rupture : des cas d'usage qui économisent des heures dès la semaine suivante.
Diagnostiquer mon adoption IAIA PME : un cadre, pas un outil de plus
Avant d'acheter une licence, on passe vos idées d'IA au filtre MAKIA (Meaning, Actors, Knowledge, Impact, Alignment). 80 % des projets « IA pour faire de l'IA » s'arrêtent là. Les 20 % restants ont un ROI démontrable en 90 jours.
Formation IA entreprise : 60 % de pratique, vos vrais dossiers
Une journée, 7 modules, 6h30 effectives sur les comptes-rendus, mails clients, appels d'offres et documentation technique de vos équipes. On quitte la salle avec des prompts qui tournent — pas avec un PDF de plus.
Outils choisis, pas empilés
ChatGPT, Claude, Gemini, NotebookLM, Notion AI, Otter. On vous apprend surtout à choisir entre eux selon le cas d'usage, et à rester souverain sur vos données (RGPD intégré dès le cadrage).
Résultats observés sur 30 PME industrielles
−30 %
sur certains coûts opérationnels
+15 %
sur les activités génératrices de revenus
+5 h
libérées par manager / semaine
Moyennes observées par Makia Labs sur 30 PME industrielles (2024–2026). Variation selon le degré d'adoption.
FREE TOOL
Your AI SME checklist: 10 questions before you invest
Download a practical checklist built on the MAKIA framework. Every box ticked = one risk avoided and a budget better spent.
- —10 concrete questions (Meaning, Actors, Knowledge, Impact, Alignment)
- —Quick score: know whether you're ready to launch
- —GDPR, governance, test before buying: nothing is forgotten
- —Printable text file to share with your team
WHITEPAPER · FREE
Generative AI in SMEs: what works, what doesn't.
In February 2025, AI training became a legal duty across the EU. Most SMEs haven't noticed. This 32-page playbook explains what changed, what to do about it, and how the MAKIA framework filters the 80% of AI projects that don't belong in your business from the 20% that do.
FAQ
Questions fréquentes
Qu'est-ce que le cadre MAKIA ?
Cinq questions posées avant tout déploiement IA : Meaning, Actors, Knowledge, Impact, Alignment. Le filtre élimine 80 % des projets « IA pour faire de l'IA » et garde les 20 % qui créent vraiment de la valeur.
Combien de temps dure une formation IA Makia ?
De ½ à 5 jours selon le niveau et la profondeur. La plupart des PME démarrent par 1 journée intensive (7 modules, 6h30 effectives, 60 % de pratique).
ChatGPT est-il conforme RGPD pour une entreprise française ?
Oui, à condition de cadrer l'usage : version Team ou Enterprise (pas la version perso), désactivation de l'entraînement sur les conversations, charte d'usage interne, et exclusion des données sensibles. C'est exactement ce que la formation Makia met en place.
En quoi c'est différent d'une formation IA générique ?
60 % du temps, les participants travaillent sur leurs vrais dossiers (comptes-rendus, mails, appels d'offres). Pas de démo. Et chaque cas d'usage passe par le filtre MAKIA pour ne pas perdre de temps.
Quels résultats attendre ?
Moyennes observées sur 30 PME industrielles (2024–2026) : −30 % sur certains coûts opérationnels, +15 % sur les activités génératrices de CA, +5 h libérées par manager et par semaine. Variation selon le degré d'adoption.
Let's talk about your AI adoption.
30 minutes. No slides. We look at one of your use cases and tell you what we'd do.