How a Solo Founder turned a 10-minute Idea into a $2300 MRR

Hey guys, how are you doing?

Do you remember the bulb guy who kept trying 999 times to make the bulb? A founder has followed in his footsteps. After dozens of failed prototypes and low-signup experiments, this founder unlocked a smarter way to let AI do market research. Using Anthropic’s Claude, they spent just 10 minutes scanning Reddit, Twitter, Quora, and Indie Hackers for freelancer pain points.

What surfaced led to PortfolioWebsite.ai, a SaaS tool that now earns $2300 per month from freelancers who want quick, clean portfolio sites without coding. All this was built without surveys or landing pages.

TL; DR

·       A solo Founder used Claude AI for instant market research instead of surveys

·       Built PortfolioWebsite.ao, a tool to auto-generate freelancer portfolios.

·       Reached $2.3k MRR within weeks, mostly via word-of-mouth.

Background

            Before PortfolioWebsite.ai was even a thing, the founder had what many solo builders have, a Notion graveyard filled with half-baked SaaS ideas. Dozens of them. Some were promising, others were pure “what if…” daydreams. But the biggest problem? No users, no interviews, and no real-world feedback. Just vibes and too much coffee.

Their early projects didn’t exactly break the internet either. One prototype earned a modest three signups (and we’re pretty sure one of those was their cousin). Another idea didn’t even make it online, it died quietly on their local machine like a shy artist refusing to perform.

That’s when something clicked. Instead of building in a vacuum again, they got smart. They turned to Claude, not the guy down the street, but the AI assistant. Using Claude’s research skills, they had it comb through forums, threads, and discussions to find actual freelancer pain points about building portfolios.

And just like that, clarity appeared. PortfolioWebsite.ai wasn’t born from guessing, it came from listening at scale, powered by AI. One specific pain point kept showing up again and again: freelancers hated building portfolios. Too much work, too little time, and no good tools. And that insight became the heartbeat of the product.

TL; DR

·       Founder had many SaaS ideas but zero clarity on Validation.

·       Traditional validation (talking to users) felt ineffective without reach

·       Claude’s “research Model” opened a fast path to demand discovery.

Challenges

            Now, if you thought, 'Well, everyone faces challenges,' you would be right. Many founders fall into the trap of building based on assumptions. That’s what happened here: products were built then abandoned due to a lack of interest. Without existing reach, user interviews were impossible. A paid ad campaign felt expensive, and gathering feedback through forms felt forced. So, the challenge wasn’t technical; it was finding real demand without wasting time.

            The Key to solving this problem was to turn Claude into a personal market-research engine. They activated research mode in Claude for deeper analysis, asked it to analyze Reddit, Twitter, Quora, Indie Hackers threads, and to extract recurring quotes like;

·       I hate coding portfolio sites from scratch

·       I don’t know what to put in my portfolio

·       I want a clean link for clients without the hassle

Results

            Turns out, listening to real problems works. Just two weeks after launching PortfolioWebsite.ai, the founder landed their first paying customer, and no, it wasn’t a friend or a pity subscriber. It was an actual freelancer who needed the tool and happily paid for it.

Things picked up quickly from there. Without any fancy marketing campaign or viral Twitter thread, the product grew through referrals and community word-of-mouth. Freelancers who used the platform shared their shiny new portfolios online, and when others asked, “Hey, how’d you build that?”, they pointed right back to the site.

Before long, the product was pulling in $2,300 in monthly recurring revenue (MRR). No paid ads. No growth hacking tricks. Just a tool that quietly solved a loud problem — and users who did the promoting without being asked.

It wasn’t flashy. But it was honest. And it worked.

TL; DR

·       They got their first paying user just two weeks post-launch.

·       Monthly revenue reached $2300 MRR within weeks.

·       Growth via product virality and earned shares.

 Ok guys, I hope you enjoyed this story and I’ll catch y’all next time.

Best Regards,

The Arcadia Studio Team