Livrer un outil IA à 5 000 € à une PME avec Claude Code (De A à Z)

Value of the Information & Strength of the Argument

The video delivers practical value by breaking down a complex process into actionable steps, emphasizing that the hardest part is not coding but finding clients and managing the project. The argumentation is persuasive, using real examples of earnings from his community members to motivate viewers. He supports his claims with a structured methodology and distinguishes between what is in his interest and what is best for the client, which adds credibility. However, the discussion is heavily oriented toward promoting his paid coaching program, and the earnings testimonials are not independently verifiable, which weakens the overall argumentation.

Scientific Rigor, Source Quality, Title Accuracy

The video is not a scientific presentation but a practical business guide. The sources cited are limited to a promotional link in the description; the mentioned tools (GitHub, Supabase, Vercel, Claude Code) are well-known platforms but are referenced without official documentation. The creator gives no external citations or data to back up his claims, relying on personal experience and testimonials. The title is well-aligned with the content, which is instructional and focused on the deliverable process. The rigour is moderate for a tutorial, with an appropriate caveat about legal and contractual matters. There are no public comments to analyse.

The title accurately reflects the content: it demonstrates the complete process of delivering an AI tool for around €5,000 to an SME using Claude Code, from customer meeting to delivery and billing.

6/10

The video provides a practical step-by-step methodology for building and delivering AI tools to SMEs, based on the creator's experience and testimonials from his community. However, the claims about earnings are anecdotal and not independently verified, and the content includes promotional segments for his coaching program. The technical explanations are simplified but coherent, and the advice on legal and contractual aspects is appropriately cautious.

Markers derived by PSI from the transcript: the creator did not define chapters.

• 02:00Introduction: the problem of non-developers delivering AI tools and the promise to show the full process.
• 05:00Step 1: Ideation using the call transcript to brainstorm the solution with Claude.
• 07:00Step 2: Creating an audit report and PRD (Product Requirements Document) to frame the project.
• 10:00Step 3: Generating CLAUDE.md from the PRD to serve as a context file for Claude Code.
• 12:00Step 4: Building the MVP using Claude Code, then introducing the three deployment tools: GitHub, Supabase, Vercel.
• 16:00Two billing models: retaining ownership (model 1) vs. transferring everything to the client (model 2), with a recommendation.
• 20:00Detailed eight-step delivery process, including contracting, RGPD considerations, and delivery via Loom video and documentation.

• Wolph Lab - AI consultant coaching — Promotional link for AI consultant training and community, mentioned throughout the video.