After five failed products, this founder quit his job and hit $7.5k/mo in 12 months

Your AI assistant can do more than answer questions.

Some marketers use Claude to write emails. The smartest ones use it to build campaigns, create strategies, and handle work that normally takes an entire team.

Get 60 proven Claude hacks that help you plan marketing campaigns, create content faster, automate workflows, and work like a full-service agency—without the agency cost.

Seven years of building on the side. Five-plus shipped projects. One hundred dollars in total revenue. On paper, a case study in what not to do.

Filip Panoski is a self-taught software developer who lived that exact record before anything changed. What makes his story worth studying is not that he eventually succeeded — it is that he treated each of those seven years of failure as a specific lesson, and then built his next product to avoid every single one of them.

He quit a $7k per month development job to go all-in, running entirely on savings with no income. That bet had to work. It became Bazzly, a Reddit marketing tool for SaaS founders, which he grew from $0 to $7.5k per month in twelve months — $5k in MRR plus $2.5k in monthly expansion revenue from one-time top-ups inside the platform.

Here is how the failures shaped the product, and the unusually specific playbook that got him there.

Filip did not pick Bazzly's shape at random. Each defining decision was a direct correction of a past failure.

One failed project was a B2C productivity tool. Bootstrapping B2C is hard — lower lifetime value, higher churn, more expensive acquisition — so this time he chose B2B. Another failure was a B2B launch platform built on one-time payments, a model that is brutal to bootstrap without large distribution because you start from zero every single day. So he committed to B2B recurring revenue solving a recurring pain point.

The niche selection was equally deliberate. He noticed Reddit marketing tools already making money, which validated demand. He knew the space firsthand from distributing his own earlier projects there. And there were no giant competitors making the market hard to enter. Real demand, personal knowledge, no dominant incumbent — the combination that makes a niche workable for a

This is the move that most separates Filip's approach from the standard path, and he is emphatic about it.

Before writing a line of code, he built a waitlist landing page and started sending DMs. The logic was a test disguised as a launch: if he could get people onto the waitlist through direct outreach, it meant he could keep acquiring customers the same way once the product existed. He got 30 people on the waitlist, and only then started building the MVP.

With the demand signal confirmed, he shipped the MVP within one month — a hard correction from previous projects where he had spent months and years building before launching. As a technical founder, AI gave him the speed to ship fast without sacrificing quality. Early running costs were near zero: free Supabase and Vercel plans, cheap or free tiers for everything else. His stack was simply what he knew best — Next.js on the frontend, Supabase for the database, Node.js on the backend. His view on stack choice is refreshingly unfussy: use what you are most comfortable with.

Bazzly runs on a subscription model with one-time top-ups, and users can pay to add project seats as they grow their own portfolio. But the pricing did not start where it landed.

He began with three plans: $19, $39, and $99. Then he eliminated the cheaper plans, repositioned around a single $99 plan, and revenue increased significantly. The one-time credit top-ups, meanwhile, became a surprisingly strong revenue lever on their own.

His B2B pricing playbook, which he says he ran twice — from $100 to $1k, then $1k to $5k — is worth stating plainly: start with a single plan that includes everything. Expand revenue through top-ups for more usage or additional seats. If the MVP is simple, start at a minimum of $39 per month. Then, as the core offer becomes more valuable, increase the price and reposition around the new value. Simplicity first, price increases as the value grows.

Filip breaks the twelve months into three distinct phases, each with a different bottleneck.

The repositioning detail matters as much as the partnership. Same product core, same distribution channels — but a sharper promise moved the revenue. The offer, not the traffic, was the constraint.

His acquisition mechanics are specific enough to copy directly.

First customers came from DMs on X and Reddit. He found people describing a problem Bazzly solves — or explicitly asking for a product like it — and sent a personalized message following one template: a quick summary of what they had posted, a single sentence tying his tool to their problem, and an offer to try it. His reasoning: DMs open one-to-one conversations that produce qualitative feedback and relationships, and they are the easiest way to get someone to try a product when you have zero social proof.

Next came high-intent Reddit posts with genuinely helpful replies that mentioned his product. Interested readers would Google the product name and land on his site. This channel compounded unusually well: Reddit posts rank on Google, the comments drove effortless monthly traffic, and he even began seeing AI mentions from ChatGPT and Perplexity that traced back to those same Reddit comments. As he puts it, he used Bazzly to grow Bazzly.

Only once he had real wins and proven strategies did he start building in public on X, Indie Hackers, and Reddit. Because the content came from genuine experience, it resonated and kept driving organic traffic. His discipline throughout: systematize every channel that works before adding a new one. Reddit brought in 50% of signups at one point; as his X audience grows, it is becoming a strong channel too, and SEO is next.

Filip's single most contrarian claim is that expanding distribution is overrated. The highest-leverage growth move, in his experience, was not adding channels — it was improving the funnel so existing channels convert better.

The numbers make the case. In January, his funnel ran roughly 10.7% visitor-to-signup, 29.6% signup-to-trial, and 30.8% trial-to-paid. By June, signup-to-trial had climbed to 35.9% and trial-to-paid to 52.6%. Even with visitor-to-signup slightly down, the same marketing effort produced nearly double the growth. The improvement came entirely from fixing conversion, not from finding new traffic.

His resulting advice: track your funnel metrics religiously — visitors, visitor-to-signup, signup-to-trial, trial-to-paid, and churn — then find the biggest bottleneck and fix that first before doing anything else.

Filip is candid that entrepreneurship is over-romanticized — the hardest thing he has ever done, full of stress and sacrifice, and not for everyone. But for anyone who feels called to it anyway, he offers a set of operating principles earned across seven years:

His closing note is simple and hard-earned: if the work calls to you, do not give up, no matter how long it gets. Stay consistent, and never stop iterating.

Filip is all-in on Bazzly. The next goal is $10k MRR and building the best Reddit marketing tool on the market — and he suspects the untapped potential in the Reddit niche alone could carry him past that faster than expected. Beyond that, he sees Bazzly expanding from a Reddit tool into a broader AI system that brings qualified traffic to SaaS companies.

You can check out Bazzly directly, and follow Filip's build-in-public journey on X.

Some people use Claude to write emails. Others use it to basically run their entire business while they play Wordle.

This isn't just ChatGPT's cooler cousin. It's the AI that's quietly revolutionizing how smart people work – writing entire business plans, planning marketing campaigns, and basically becoming the intern you never have to pay.

The Hustle's new guide shows you exactly how the AI-literate are leaving everyone else behind. Subscribe for instant access.

Get our Claude playbook