Introduction

This is a new series where I’m documenting the entire process of building my 6th SaaS product.

I’ve launched 5 SaaS products in previous years. These were:

szabaduszok.com – (szabaduszok means freelancers in my language) A freelancer platform for the Hungarian market. It was similar to freelancer.com or Fiverr. It had about 3,000 free users. Never had paying customers, but I sold the project for a small amount of money.
Laracheck – A code review tool for Laravel projects. My favorite feature was N+1 query detection. It still has 3 or 4 paying customers.
junior.dev – Code generation tool for Laravel projects. It could generate full CRUD functionalities and tests. Never monetized it.
ReelCode – A LeetCode alternative with more practical challenges. It had a free and a premium plan. Didn’t have significant revenue.
ContentSoup – A content remix tool. You give an article to it, and it creates 25 Twitter threads, etc. I mainly built it for myself and sent it to some content creators. It wasn’t successful, and I didn’t want to invest a lot of time and money to reach out to content creators.
deQenQ – A software development company where we focus on system design, performance optimization, and AI integration. This is not a SaaS product, but I wanted to include it.

Most of them failed without significant revenue.

So this is not a series where I tell you how to launch a successful SaaS.

Because, clearly, I have no idea.

This is me documenting my 6th SaaS adventure.

Reflecting on my failures

I’m 99% sure what went wrong with my previous projects. It’s one word.

Marketing.

Building a SaaS, for me, and for lots of developers, looks like this:

70% of coding
20% of agonizing over the deployment process
10% of marketing

But it probably should look like this:

1% of coding
99% of selling it to real people for real money

This is bad news for us, developers.

For this product, I’ll follow some ideas from the book “Traction.” This one, to be more precise:

From day 1, I’ll dedicate 50% of my time to product and 50% to marketing.

I’ll share lots of things about my marketing approach in the following posts. But first, let’s talk about the idea.

The idea

Where is my docker volume?

It was the 15th of August, around 9AM. I was about to deploy a new project and set up HTTPS.

The app was running in a docker compose stack in production. I wanted to add certbot.

I remembered that certbot has a Docker image and that I need to mount some directories from the proxy to certbot or something like that. I’ve already done this a few times in the past. But I had no idea when, or in which project, etc.

Of course, I couldn’t find anything.

So I started the entire process from scratch. Again. Probably for the 4th time in my life. It took like an hour, it was clunky, I hated my life.

Why the hell am I not able to ask an application a question like: in which project did I do the exact same thing and how?

Who the hell is Peter?

It was 3:30PM when my calendar sent a notification:

You have an event in 30 minutes, 4:00PM “Peter x Martin”

The meeting didn’t have additional context.

Who the hell is Peter and what am I supposed to talk about?

If you think I’m lame, yes I am, but on top of that I have ~1,100 Twitter conversations from the last ~40 days:

I start conversations with my followers and offer a 30-60min Google Meet for anyone who needs a little help. I don’t know my followers personally, so there can be lots of potential “Peters.”

I quickly started searching in emails, Twitter messages, Drive memos, etc.

After ~15 minutes, I found out who Peter was.

Why the hell am I not able to ask an application a question like: please remind me who Peter is and what’s the topic of our 4PM meeting?

Thanks to certbot and Peter, I started thinking about an app that solves these kinds of problems (and man, I have lots of problems like these).