Back

On starting a SaaS

February 6th, 2025

Roughly three quarters of a year ago, I was so frustrated working with [redacted] that I decided this could be done better. The tool I had to deal with was a 'all-in-one' solution for churches, providing customers with a website, app, some community tools and a bunch of other related stuff. Almost every feature was broken in some way, and when reporting these bugs, even with full-on reproduction videos, they were basically ignored.

This convinced me to start building Kerkcloud. I think the journey to where I am now, a first release, is worth sharing. I will write it in chronological order, starting with the idea and ending with the release.

The journey

The idea

I had a realisation: I can build that, and I can do a way better job. And for sure, it won't have to cost 150 euros a month. So I went to the drawing board, checking what would be important requirements and how to set such a project up.

The classic developer trap

Yeah I can build that, I did build it in fact, but that doesn't mean anyone will use it. But first I had to go about it like many developers before me: build build build.

  1. Make a POC, see if the core idea is feasible
  2. Build a MVP, see if people are interested
  3. Keep moving the goal posts and never finish the MVP

This went from "Soon I'll get some people on board to try it, then iterate", to "Just a few more features. I'll have more time to finish it the coming months (not true, ever)", to finally "I fell for the classic developer trap".

Sunk cost fallacy

Sounds bad, but saved the project. Around half a year in, I was honest with myself about not actually making enough time to ever release something. Even though I learned a lot and had a lot of fun building it, I felt that I was still losing something if I gave up now. This pushed me on to make more time and actually finish the MVP.

Code done, devops?

Finally I set clearer and clearer boundaries, making decisions to delay features where possible and focus on the core. This meant I wrapped up a working and production ready version in two more months. Now I had to actually deploy it, properly.

I thought this would take a few days, but I think I spent a hundred hours on it during Christmas break. Cutting corners is just not an options here, and I'm glad I didn't. But that was a blow, going back to work and university without having time to actually launch it.

Launch

What some would call a launch, has been a part-time exerice of two-hour blocks at a time. Little spare time once more and many other important things going on. But it is actually live now! After ten months of on and off work, something works.

What now?

Marketing and sales

Hah. This is a first for me. I have plenty of experience in building web applications, but it doesn't sell itself unfortunately. So this is a first for me! I am on a journey to learn how to market and sell a product. I have no idea what I'm doing, but I'm doing it anyway (Do reach out if you have recommendations!).

Features

I want to keep building, so as soon as I get a clear picture of what my (potential) customers want, I will put them at the top of my list and start pushing code once more! Plenty of things in the pipeline, but the focus is on building the right things.

Conclusion

This has been quite the journey, and one that is hopefully just getting started. I hope to share more in the future, also more technical details and insights. If you have any questions, feel free to reach out via e-mail. Interested in the product? Check out Kerkcloud.