How I Built My First SaaS in 30 Days

The 30-Day Challenge

Most founders spend months (or years) polished a product that nobody wants. I decided to do the opposite: ship a messy MVP in 30 days and see if it sticks.

Here's exactly how I did it.

Week 1: Idea & Validation

I didn't start with code. I started with a problem I had myself: creating decent-looking blog cover images took way too long.

I posted on Twitter: "Building a tool to generate blog covers from headlines automatically. Who wants beta access?"

I got 50 signups in 24 hours. Validation: ✅

Week 2: The "Boring" Stack

I chose the stack I knew best. No new frameworks. No fancy graph databases.

Frontend: Next.js (hosted on Vercel)
Backend: Next.js API Routes
Database: Supabase (Postgres)
Auth: Supabase Auth
AI: OpenAI API

Lesson: Innovation tokens are for features, not your tech stack. Use what you know.

Week 3: Building the Core Feature

I focused on one thing: input headline -> get image.

I stripped everything else.

No user profiles (just email login)
No billing dashboard (just a Stripe link)
No history (save images locally)

It was ugly, but it worked.

Week 4: Launch & First Revenue

I deployed on a Tuesday. I emailed my 50 beta users.

"It's live. It's $9/month. Here's a 50% off code for early birds."

By Friday, I had my first 3 paying customers. $27 MRR. It wasn't much, but it was infinite growth from $0.

Key Takeaways

Stop overthinking. Your first version will suck. Ship it anyway.
Validate early. If I hadn't gotten those 50 signups, I wouldn't have built it.
Charge immediately. Free users give feedback. Paid users give validation.

What's Next?

Now comes the hard part: marketing and retention. But the product exists, and people are paying for it. That's a win.

Want to see the tool in action? Check out PublishPix.