Loading page…
An AI business coach packages a founder’s client-acquisition playbook into a paid digital adviser for freelancers and aspiring service founders.
Aug. 12–Nov. 24 (about 3 months)
Validate the advice before building the software: Bruno sold a custom GPT version first, then used buyer feedback and launch economics to justify Agentor.
Fewer bars = easier, cheaper, or faster for an AI-assisted solo builder. Editorial judgments based on the case details.
Bruno Gabarra started with a familiar problem among digital service providers: freelancers, traffic managers, copywriters, designers and small AI-service operators struggled to find clients. He had previously worked as a traffic manager and agency owner, and had a sales method he could teach. Instead of putting that method into another course, he first packaged it as a custom GPT called Gabarra AI, a conversational adviser trained to deliver his approach. He says buyers responded strongly because they could ask for guidance whenever they needed it. That early paid validation led him to build a web application with Lovable, using Claude to review the code. The product, later named Agentor, began as a client-getting coach and expanded into guidance on launching digital services, sales, funnels, paid traffic, copywriting and AI agents. Gabarra reports R$449,687 in sales from Aug. 12 to Nov. 24, a period he describes as roughly three months. He says an earlier validation phase generated another R$30,000, outside that period. The headline figure is a founder claim tied to a payment-platform screenshot he describes, not audited accounts.
For an AI adviser aimed at freelancers, start where the buyer already spends time learning how to win work: a live workshop, a niche creator audience, or a practical community for designers, traffic managers and copywriters.
The quick-reference card — acquisition channels, replication playbook, and risk map — is members-only.
Join the Unicorn Club — $5/moCancel anytime, in one clickPaste into Claude Code / Codex and get a working version of this product end-to-end.
Two production-grade prompts per idea — a full build spec and an SEO growth plan — plus 10 tailored build specs a week. Members only.
Join the Unicorn Club — $5/moCancel anytime, in one clickEvery claim traces back to a listed source — open any of them and check the number yourself.
The full captured transcripts behind these sources are in the Unicorn Club — search them, quote them, check every figure. Join the Unicorn Club
Data credibility: Bruno Gabarra reports R$449,687 in sales for Aug. 12–Nov. 24 and describes showing a payment-platform screenshot. The figure is creator-reported; the transcript provides no independent verification, transaction export, customer count, price breakdown, refunds, or recurring revenue. He separately reports R$30,000 in an earlier validation phase. The stated development and monthly software costs are also founder-reported.
The first sales came from an online class and a paid GPT assistant, not from a finished subscription application. Gabarra says he invested R$5,000 in validating the offer and generated R$30,000 in sales during that earlier phase; he also says the amount was an affordable risk for him, not a universal startup budget.
Agentor’s growth loop began with a repeatable launch, then improved as the product incorporated more of the founder’s expertise. Gabarra says he repeated and refined the online launch after validating the idea, reinvesting a portion of proceeds into customer acquisition.
An early knowledge product can feel quiet between launches: buyers may use the adviser privately, and many will not volunteer feedback unless asked at the moment they hit friction. Gabarra’s approach reduced the cost of that uncertainty by selling a lightweight custom GPT before building the full application.
The rest of this breakdown — 9 more sections — is in the Unicorn Club, with every section of all 1,012 ideas: first customers, first dollar, the flywheel, and the silent stretch.
Join the Unicorn Club — $5/moCancel anytime, in one click