White paper no. 1
AI Maturity of French SMEs 2025-2026
Where do you really stand? The Junyr Method™ Scale in 5 levels, the 5 fatal mistakes, sovereignty & FinOps, the INDUSTEC case study (182% ROI in 9 months) and a 12-month roadmap.
58% of SMEs use AI.
Only 11% get real value from it.
I work with the 11%.
AI in SMEs is not won through technology. It is won through method and change management.
AI in the service of people.
AI project lead, AI transformation practitioner trained on the Junyr Method™. Fifteen years of transition management · 30+ documented engagements with French SMEs and mid-caps · sovereign infrastructure. From strategic diagnosis to operational deployments, for business leaders aiming at a structural competitive advantage, not scattergun adoption.
Free audit · 30 minutes by video conference · no obligation.
My approach, on video · 7 min
Subtitles available in English and French ·
Also online (in French): La Martingale #328 · “Investing in AI: how to spot the companies that will truly create value” ↗ (30 July 2026). 15+ years
of transition management
30+ engagements
documented with SMEs and mid-caps
159.8%
median documented ROI over 24 months, well-framed AI projects
5 levels
on the Junyr Method™ Scale, none can be skipped
Source: Baromètre IA & ROI PME France 2022-2025, Denis Atlan (CC BY 4.0)
Career
Thirty years of practice, six milestones
The job has never changed: taking the helm of an organisation at the moment it must change, and making that change succeed. Diagnose, frame, execute, measure, hand over.
- 1993
First lines of code
A school-management system, image processing for the French Ministry of Defence, a genetic algorithm in operations research: eight years of practice before the engineering degree.
- 2001
Engineer, École des Mines de Nantes
Graduated in Organisation and Management of Information Technologies. The school formalised what eight years of practice had built: moving from code to strategy without changing language.
- 2001-2003
Bureau Veritas, e-Business Consultant
Contribution under the aegis of the United Nations Economic Commission for Europe (Geneva) on regulating e-marketplaces through a trusted third party.
- 2005-2021
Fifteen years of transition management
Seven sectors, more than thirty engagements: recruitment (RhFed, 150+ hires), education and research (ITEMM, Sorbonne Foundation), real estate, events, specialised retail.
- 2023
The AI pivot
Designing and operating orchestrated AI agent systems, run on fully sovereign infrastructure. In parallel, a law degree at Université Paris-Sorbonne (2023-2026): the legal framework alongside code and strategy.
- 2026
The Junyr ecosystem
Launch of Junyr Suite (junyr.app): the tool first built for engagements, now made available. The Junyr.eu network of AI transformation practitioners. And a 100% solar-powered datacenter dedicated to AI.
The method
Five maturity levels, none can be skipped
N1 Spectateur
N2 Artisan
N3 Orchestre
N4 Architecte
N5 Pionnier
The Junyr Method™ was born of fifteen years of transition management and more than thirty documented engagements with SMEs. It places your company on five rungs and charts the pragmatic path to climb them one rung at a time. The key step, from Artisan to Orchestre, is within reach of any SME or mid-cap.
The golden rule: no level can be skipped. An Artisan-level SME that invests €200,000 to build an autonomous agent platform fails more than one time in two: not for lack of technology, but for lack of foundations.
White papers
Find your position, industrialise, then hold up
In Q2 2026, 58% of French SMEs and micro-businesses use AI, nearly half of them daily. Yet only ≈11% derive measured value from it, and 43% of AI initiatives are still judged doomed to fail by their own decision-makers. These three volumes document the path, unvarnished: the first positions you, the second sets you in motion, the third makes what your agents execute hold up.
White paper no. 2
From PoC to Industrialisation
How do you industrialise? The 7 frictions of the “last mile”, the 5 levers of change management, AgentOps, ROI and a 12-month roadmap. FR + EN edition, June 2026.
White paper no. 3
Agentic processes and tomorrow's interfaces
How does it hold up? Skill versus agentic workflow, the reliability cost of nondeterminism, the four software engineering gestures, the control room and the three tiers of use. FR + EN edition, August 2026.
The blog
39 AI insights, organised by theme
AI strategy, autonomous agents, digital sovereignty, FinOps, cybersecurity. Published at the pace of the field, in French and in English.
Junyr Method™ 12 articles
- 29 July 2026 Decision-Gated Development: when the bottleneck moves from execution to decision
- 15 July 2026 We removed Git from a production product: roadmap-driven development
- 30 June 2026 How to involve employee representative bodies (IRP) in your AI transformation: what recent case law says
- 23 June 2026 SaaS & AI: from technical debt to orchestration
- 23 June 2026 AI & promotional products: the anti-disintermediation playbook
- 23 June 2026 Asynchronous work: in the age of AI agents, everyone becomes a manager
- 23 June 2026 Vibe coding is dead: enter UltraCoding
- 14 June 2026 Junyr: you are not buying an AI. You are buying the base your AI works on.
- 7 June 2026 Where is AI's value heading? Three signals of commoditisation that business leaders must read before the markets do
- 1 June 2026 Junyr Agents™: delegating AI in your SME without losing control
- 29 May 2026 Cryptography 2026: transport, storage, identities and the post-quantum transition, the decision framework for business leaders and CISOs
- 23 May 2026 The Junyr Method™ Scale: the 5 AI maturity levels of an SME (and why none can be skipped)
Transition Management 9 articles
- 26 June 2026 The end of email and Office suites by 2030?
- 31 May 2026 The end of the telephone: why the chat + video pairing is becoming the standard for professional communication in 2026
- 19 May 2026 The end of prompt engineering: why your teams must stop talking to AI and start commanding it
- 5 November 2025 From win-win to ethical hard bargaining: why I had to change my negotiation method
- 1 September 2025 Alternative Dispute Resolution (ADR) in the Service of Organisational Performance
- 19 November 2024 AI in micro-businesses: a revolution similar to 1980s computing?
- 16 June 2024 Supporting the internal promotion of a sales manager in a communications agency
- 9 June 2024 Implementing artificial intelligence to grow a communications agency
- 2 June 2024 Recruiting Generation Z: innovative strategies to attract tomorrow's talent
AI Agents 6 articles
- 30 August 2026 Enterprise agentic processes: a skill is reused, a workflow is replayed
- 27 August 2026 Agentic design: why Cowork does not build durable agents
- 27 August 2026 Conversational AI cannot become a work interface
- 18 August 2026 Decide with your data, not your habits: 360° analysis now costs next to nothing
- 12 July 2026 Optimising context window size in LLMs: between cognitive dilution and hardware constraints
- 4 July 2026 Engineering agentic systems: golden rules, architecture and security of the “Human-in-the-Loop”
Markets & AI 5 articles
- 17 August 2026 The end of the model race: frontier AI is becoming a commodity
- 2 August 2026 Saaspocalypse: why 10,000 new pieces of software a day is not an exaggeration
- 23 July 2026 AI's Speculative Bubble: the Microsoft-Mistral Sovereign Pivot
- 20 July 2026 CAC 40 and Nasdaq Top 10: size does not tell you AI maturity
- 20 July 2026 The cash of AI: who pockets it, and why it may be circular
Personal Effectiveness 4 articles
- 28 August 2026 The dash and the balance: anatomy of the AI writing signature
- 7 August 2026 Cutting your mental load with AI: the personal-effectiveness method
- 2 June 2026 Video prospecting in 2026: the standard for a first meeting that converts
- 18 May 2026 Token and AI API budgets: the FinOps guide for SMEs in 2026
Case study 1 article
Cybersecurity 1 article
Methodology 1 article
FAQ
Frequently asked questions
- What is an AI project lead?
- An operational leader who takes charge of an SME or mid-cap's AI transformation for a defined period: diagnosis, framing, execution, measurement, handover. As an AI transformation practitioner trained on the Junyr Method™, I work with companies of 50 to 500 staff, with a dual stance: a developer who leads, a leader who codes.
- What is the Junyr Method™?
- A proprietary AI-transformation methodology built on fifteen years of transition management and more than thirty documented engagements. It places the company on five maturity levels (Spectateur, Artisan, Orchestre, Architecte, Pionnier) and charts the path to climb them one rung at a time, skipping none.
- How much does an AI transformation cost in an SME?
- The key step, from Artisan to Orchestre, represents an investment of €30,000 to €80,000 over 12 months, support and licences included, excluding internal time. Well-framed projects document a median ROI of 159.8% over 24 months, with a 73% success rate. Source: AI & ROI Barometer for French SMEs 2022-2025, Denis Atlan (CC BY 4.0)
- What is the AI maturity audit?
- 30 minutes by video call, free of charge and with no obligation, led by a Junyr.eu network practitioner: where you stand on the Junyr Method™ Scale, what is blocking the next level, and the first initiative worth pursuing. A one-page takeaway is sent after the session.
- Do you need sovereign infrastructure to deploy AI?
- It is an option to weigh against your data and regulatory exposure, not a dogma: SecNumCloud-qualified trusted cloud, or on-premise deployment when the context demands it. The Junyr Method™ builds this trade-off into the framing stage.
Where to start
The AI maturity audit
30 minutes by video call, led by a Junyr.eu network practitioner. You leave with your position on the Junyr Method™ Scale, what is blocking the next level, and the first initiative worth pursuing. A one-page takeaway is sent after the session.
No obligation. The audit is free of charge.