From 600 Days of Failure to $1.3k MRR
How AI turned me into a one-person factory (and what it cost me to learn this)
Two years ago, I quit my job without a plan. I thought I’d build products, launch them, and people would just show up.
They didn’t.
I wasted weeks and months building products and features I didn’t know if anybody needed. I obsessed over landing pages and copy, when I didn’t have a single user.
WriteStack was the product that snapped me out of that.
I didn’t build it on a whim or based on someone else’s ideas that I thought were valuable.
It came from a pain. Pain people told me about, not pain that I thought they felt.
Staying consistent with Notes.
Substack had no Taplio-like tool. No way to schedule Notes. No system to keep you showing up daily without burning out. And I thought: “There has to be one.”
So I built it. And AI made it possible for me to do it alone, fast, and in public.
Killing the “Perfect Tool” Illusion
When I started, I thought success meant building the perfect tool.
I thought I should just build as many products as I can, as fast as I can, until I hit a product that will go viral and make me millions.
And every product had to be flawless, with as little bugs as possible and have the best UX.
Well.. That is very far from the truth.
All the projects I built and shipped
I built over 12 well polished products. I had maybe 60 non-paying users, and most came from one product.
So I decided to go about it in a different way.
I’ll build a product, ship it as fast as I can, find users to try it and give feedback and improve.
And I gave myself 6 months to do it, or I am quitting everything all together.
Boy, was that one of the best decisions I made. I realized it doesn’t matter if your product is complete or not. Whether it’s a little buggy, not UX-optimized or if the copy is bad.
What matters is if there’s a hungry crowd for it.
And AI helps me take action on my decisions and ideas extremely fast, and test them to see how users react.
The One-Person Factory
The first version of WriteStack wasn’t fancy. It was a basic, Substack-like text editor, with a little AI on top to help generate article ideas and outlines.
It took me about a week to ship.
With Cursor + Claude, I could build the base of any feature in a single day. Here’s my flow:
I explain shortly what I want to the Chat → it helps me build an elaborated prompt → I send it Cursor to build the feature.
That’s why I call WriteStack a one-person factory. AI lets me work like a small team.
Want a new dashboard view? First 5 minutes: Claude builds the skeleton. Next few hours: I debug, fix and ship.
Need copy for the landing page? GPT writes several options, I pick the one that sounds best and tweak them, share online and get feedback.
The compounding effect is crazy. I could literally ship a very complex feature in a few days (or less).
But not every AI-powered shortcut was a win. My biggest mistake was building a real-time ghostwriting feature.
It sounded cool: you can have a ghost writer that’d write for you and schedule for you, and it’d be updated live.
The problem is that it was complicated, buggy, and nobody needed it live.
Polling data every minute would’ve been enough. I wasted days on it before I realized the mistake.
Just because AI makes something possible doesn’t mean it makes it useful.
From Zero to $1.3K MRR
The road from Zero to $1.3k was LONG and hard.
No, but seriously, I spent almost 600 days building crap nobody wanted. I couldn’t find a single feature that’d make someone pay.
And the number may not sound huge, but for me it was a make or break moment.
If I didn’t make anything valuable out of WriteStack, I’d have quit.
So if I leave behind all the mess I made in the first 600 days and focus on the one thing that made the whole difference, it’s this:
Image generated by ChatGPT
I read about it so much, I knew that I have to do it but I just didn’t have a clue where to start.
And the moment I decided to start, I slowly figured it out.
I started sending DMs to creators who write about non-fictional stuff, like productivity, personal growth etc.
After a few dozen DMs and negative feedback, I pivoted from Posts to Notes.
That’s when people started reacting positively and giving me more ideas and tips on how to improve it.
I knew I am on to something.
For the first time in a 600 days journey, I finally built something that people find useful.
And you know, the moment I knew that I have something truly valuable that is when I got a DM from a satisfied user.
She told me she’s a mom with full-time a job, and WriteStack frees up four hours a week for her to spend with her family.
That’s the kind of validation money can’t buy.
Of course, there were mistakes. I wasted time stacking features instead of getting feedback on the ones I already had.
AI made shipping so fast that I fell into the trap of shipping too much.
Eventually, I learned: kill what doesn’t matter. Focus on what brings value.
That’s why the Editor, which I thought was going to be a killer feature, is gone now.
How WriteStack Changed Me
Building WriteStack taught me something: people don’t come just because it solves their problem.
In the beginning, people will come because of you and then to solve their problem.
You might think:
“Orel, it doesn’t make sense. People are looking for solutions for their problem and they’re willing to pay for it”.
Maybe, But let me put it this way.
Would you pay an annual price for a product of someone you don’t know and who doesn’t have any social proof?
Of course not.
It might be just another indie hacker who vibe coded his way and will drop the product faster than he built it.
Yes, you might try it for a month, if the price is right.
But unless you the solution is perfectly tailored to your problem, you will churn within a month or two.
On the other hand, if the creator is supportive, gives you personal care and makes you feel like you’re the only user they have, how would you feel?
You’d probably be okay with sticking around, even if the solution is a bit messy.
Takeaway
Don’t chase the perfect product. Don’t stack features nobody asked for. Don’t waste months polishing what no one cares about.
The only thing that matters is whether there’s a hungry crowd and whether you’re willing to put yourself out there.
People first buy you. Then they buy a solution to their problem
Ship fast, kill what doesn’t matter, listen to users, and keep showing up.
That’s how you go from zero to something real.
P.S.
For the next 48 hours, you can try WriteStack for free and get 10% off any plan you choose.
Use the code TBB10 at the checkout









You turned failure to a valuable lesson and thats is growth
Orel, Terrific work! That's the thing about failing, it will happen, but they become wins when instead of being wasted, they are used as lessons for a better next try.