I Couldn't Scale My Agency. Then I Gave It an AI Brain.

I didn't start in marketing. I studied PR, did a Master's in Commercial Management, and spent six years selling at multinationals. I learned how businesses really work — not from a textbook, but from the inside.

Eventually, I left. I wanted freedom, I wanted to build something of my own, and I was tired of making other people's numbers. I tried online stores, marketplaces, digital ventures. They all failed. Every single one.

But each failure taught me something about digital business that I couldn't have learned any other way. And after a few years of failing forward, I found my lane: performance marketing. Meta Ads, Google Ads, funnels, landing pages. I was good at it. Clients started coming.

I built an agency. Fan Convert. We specialized in high-ticket clients in regulated industries — finance, real estate, investment education. The kind of work where one wrong ad can get your account banned, and where the difference between a good campaign and a great one is understanding buyer psychology at a deep level.

For a while, things were great. Clients were happy. Revenue was growing. I'd finally found the thing.

I'd always been interested in automation. Even back in my freelance days, I was the guy trying to connect tools, build workflows, eliminate repetitive tasks. But the technology wasn't there yet. Zapier could move data around, but it couldn't think.

Then AI got good. Not the "chat with a bot" kind of good. The kind where you could give it a client brief and get back copy that was actually usable. Where you could feed it campaign data and get strategic analysis that matched what a senior media buyer would say.

I started experimenting. Not with generic AI tools — with systems designed specifically for how my agency worked. I built an AI copywriting system adapted to Andromeda, Meta's latest algorithm update, generating ad variations that target 25 distinct buyer personas across 5 levels of awareness. I built an AI reporting system that reads campaign data and gives verdicts: scale this, kill that, test more of this.

One process at a time, my agency started changing.

The work that used to take my team days now took minutes. The copy that required three rounds of revisions came out nearly production-ready on the first pass. Reports that I used to build manually every Monday morning were generated automatically with strategic recommendations included.

I wasn't just more efficient. I was running a fundamentally different kind of agency.

At some point, I realized something: every system I was building for Fan Convert could work for any agency. The problems I was solving weren't unique to me. Every agency owner I talked to had the same frustrations — the talent bottleneck, the manual grind, the feeling of being trapped in the day-to-day.

So in mid-2025, I found two partners who saw the same opportunity. Virgil, a marketing veteran and expert copywriter who became our CEO — the person who brought us all together. And Vinod, an engineer who could turn my messy agency workflows into actual software.

Together, we started building Sucana.

The idea is simple: take every process I've built and tested inside my own agency, and turn it into an AI operating system that any agency can plug into. No coding. No months of setup. Just a system that understands how agencies actually work — because it was built inside one.

I'm the Head of Product. Which really means: I'm designing the tool I wish I'd had when I started Fan Convert. Every feature exists because it solved a real problem in my agency first.