How We Saved Asia Baldoni Fitness €500,000 with AI

The video above is the full case study: the interview with Sebastiano Tartufoli, filmed in our offices two months after the two training days. The figures below are his, said on camera. The video is in Italian.

In short: from problem to solution

• The problem: an entrepreneur already comfortable with AI, running a company of about twenty people, paying thousands of euros a month for software that still lacked the features he needed. He had been building the client app on his own for five months, and it was not usable.
• The solution: two days on site with the whole team. Day one, a second brain for every person. Day two, the skills, built with their own hands on their own processes.
• The result: software down from €5,000-6,000 a month to under €1,000, the app finished in two weeks, and half a million euros of value by year end on his own count.

The company

Metodo ABF is the system Asia Baldoni uses to help women deal with cellulite. She runs it with her husband Sebastiano Tartufoli: about twenty people between coaches, sales and content, and around 2,000 women helped so far.

This is not a company new to technology. Sebastiano calls himself a tinkerer, he was already watching tutorials at night and using AI every day. He is the kind of client who arrives convinced he already knows enough.

The problem: knowing how to use AI, not knowing where to start

The bottleneck was not one department. It was that the things that would have made the difference never got started, because there was no way to finish them. His words are precise: these were not things that used to take too long, they were things I would not have done at all.

The clearest case is the client app. He had worked on it from January to July, five months, with something in hand but nothing usable. Meanwhile he was paying $1,500-2,000 a month for the CRM alone, an English-language product his team struggled with, missing the features he kept asking for.

What we did: two days, then they carried on alone

Day one, the foundations: a second brain for every person, connected to the software they actually use. Day two, the skills, built together on real processes.

What matters is not the skills that came out of those two days. Not one survived unchanged: they rewrote all of them. What matters is the method, and the fact that they now build their own. Those are his words too, and they are why the calls with us stopped after the workshop.

What they built afterwards

• Their own CRM, replacing the one they were paying for. Inside: calls placed from the CRM itself, payments connected, KPIs tracked automatically, and an AI that reviews every salesperson's calls at the end of the week and leaves the feedback in their record.
• The client app, wired on the back end to the coaches' system: sales people can see where a client stands without asking anyone. Finished in two weeks.
• A content project management platform, where every video is a project with status, versions and per-editor stats.
• A Mac Mini always on in the office, running the routines and updating itself, set up together during the workshop. It runs even when Sebastiano's laptop is closed.

The telling part is who built them. Leonardo, who edits the videos, automated his own editing software by himself. Giulio wrote the software that runs their Instagram, without knowing a line of Python. Asia, who is not technical and during Covid struggled to join a video call, now works with Claude by dictating.

The part we did not expect

His doubt before signing was not about us. It was about himself: he thought he already knew enough, and did the two days mainly for the team. Looking back he says the opposite, that they would have been worth it for him alone.

It is the most common objection among technical founders, and the weakest: the distance between knowing how to use AI and knowing how to build on it is exactly the work of those two days.

What this case does not prove

It does not prove that everyone goes back to the office and rewrites their CRM: Sebastiano was already technical, and he spent the two months after the workshop building. Nor does it prove the half million is extra revenue: it is value, meaning cost avoided and work he would otherwise have paid someone else for.

What it does prove is narrower and more useful: two days done properly remove the block that was holding back what the company already wanted to do.

What comes next

His next step is small, vertical models running locally, one per department, on a machine in the office. Not for the sake of it: on that slice the data never leaves, and the model becomes a company asset.