I shipped 0 products in 7 years. Then I shipped 2 in 6 months. Here's what changed.

I R — Mon, 20 Apr 2026 00:26:21 +0000

For 7 years, I was the dev companies called when their app was on fire. I'd come in, fix the architecture, ship the feature, leave. Senior dev by day, "I'll start my own thing soon" by night.

The thing is, I never started it. Not really.

I had Notion docs full of ideas. Half-finished Figma files. A graveyard of npx create-next-app folders on my SSD. Every weekend I'd "start" something and Sunday night I'd close the laptop, vaguely frustrated, and open Slack on Monday for the client work.

This year I shipped two products in six months. Not perfect ones. But real ones, with real users, real Stripe webhooks firing.

I want to share what actually changed, because it's not what I expected.

It wasn't a new framework

I didn't switch to Bun. I didn't migrate to Server Components. I didn't discover some 10x productivity stack. I'm using the same stack I've used for years. The tools were never the bottleneck. I just told myself they were because tweaking a stack feels like progress without the risk of shipping.

If you're rewriting your boilerplate for the third time this year, you already know.

What actually changed: I stopped designing in Figma first

For years I'd open Figma, spend a weekend on a beautiful landing page mockup, then never write a line of code because the gap between the mockup and shippable felt too big.

Now I open the code editor first. Tailwind. shadcn. Ugly v1 in the browser within an hour. The product exists. From there I iterate on a real thing, not a fantasy.

This sounds obvious. It took me 7 years to actually do it.

I picked problems I had myself

Both products solve problems I was actively bleeding from.

The first one detects blockers across the tools I lived in every day — because I was losing 2 hours a day context-switching.

The second one helps founders find conversations where their product is the answer — because I'd just shipped the first one and discovered that building is the easy 20%. Distribution is the 80% nobody warns you about.

When you have the problem yourself, you don't need user interviews to know if the feature is right. You ship it Friday, use it Monday, and you know within a week.

I gave myself a real deadline

Not "I'll launch this quarter". A specific date, written down, told to people who'd ask me about it.

Public commitment is the cheat code. Once 5 people on Twitter know you're shipping on the 15th, you ship on the 15th. Or you look like every other indie hacker who keeps "almost launching".

The part nobody talks about

Here's what shocked me most as a dev: shipping the product was the easiest part of the journey.

I shipped my first SaaS in 6 weeks. Then I posted on Reddit. 3 upvotes. One bot comment.

That moment broke something in my dev brain. I'd built a clean codebase with proper TypeScript types, tested onboarding flows, beautiful animations — and the world genuinely did not care.

So I built a second product to solve that exact problem: finding the right conversations to join, where my actual product was the actual answer. Meta? Yes. But also the most authentic dogfooding I've ever done — I'm using it to find users for itself, and to find users for the first one.

What I'd tell my younger self

Stop optimizing the stack
Ship the ugly v1 this weekend
Pick a problem you bleed from
Tell 5 people the launch date
Start thinking about distribution the day you start coding, not the day you launch

That last one cost me about 6 years.

If you're a dev who shipped and got crickets — try Novaseed

What's the thing that's been sitting in your half-finished projects folder? Curious to hear in the comments.