What are you working on?
My project is called Amon, and it's a hosted/self-hosted server and application monitoring tool. I've been working on Amon in one shape or another since 2011, and full-time since 2014.
How'd you get started?
It started as an open-source side project, then got popular on Hacker News, and I decided to continue working on it and release a paid version.
Amon is a product for developers/devops/system administrators. My initial motivation for starting the project was to create a monitoring tool that does not stand in your way — one that's easy to install, doesn't have a steep learning curve and works out of the box.
That was an issue 5 years ago when I started, and it is still an issue today. Developer-oriented tools don't have the same level of polish as consumer products, and they rely too much on our intelligence and willingness to spend time figuring things out on our own.
Amon is completely self-funded. I spent probably 10 hours a week on the project between 2011 and 2014, and after burning out from work in 2014, I decided to take the plunge and work on Amon full-time.
What gave you the confidence to quit your job?
I've always worked on my own projects, and I've always seen my job/consulting gig as a way to support my own stuff. Amon was my first project that was successful in a commercial way.
The issue with 99% of the jobs out there is that they box you in. You are either a programmer or a designer or a marketing guy or a sales guy. I am a little bit of all these things, and I enjoy changing lanes and working on different business segments. One way to avoid getting boxed in is to work on your own project or found/co-found a company.
However, working for yourself is extremely challenging. You have to adapt and — most importantly — fail fast. Sticking to inertia almost always leads to failure.
Where'd you get the idea for Amon?
Development tools don't have the level of polish expected in consumer products. We as hackers/programmers are expected to figure out how things work on our own — to read through source code, read forums, to fix code if it is open-source, etc.
That's especially true when it comes down to monitoring. I initially got the idea when I spent a couple of days trying to install and configure a popular open-source monitoring tool called Nagios.
I looked around and saw that most open-source tools are hard to install/update, and decided that the status quo is unacceptable for me. Amon is a self-hosted tool, and my idea from the start was to have it up and running in less than 2 minutes. That is still true today.
I did consider working on other ideas, and I have a big list of potential projects, but in the last 2 years I've only worked on Amon.
What's your tech stack?
I used a very boring and predictable stack with a great ROI — Python and Django with Angular 1 and D3 on the front-end.
How and when did you start making money from Amon?
Initially I had an open-source version and a paid version with more features. People used the open-source version — if they liked it, they bought the paid version after a couple of months. My end goal was to fully support myself. That was not possible in the first year, when I made around $400-500/month.
Nowadays I want to fully transition from self-hosted to a cloud hosted tool — subscriptions and recurring revenue are much more predictable, compared to one-time license fees.
My peak monthly revenue is around $2500/mo and the average is around $1500/mo. It really depends on how much time I spend on marketing and how much on developing new features. As a single founder, that's the most difficult part to manage. I usually have one month that's development heavy, and the next month I spend more of my time on marketing, analyzing customer feedback, etc.
How did you market Amon after launching?
I write articles about things that are particularly interesting for me in the Monitoring/DevOps space and I share them on Hacker News, Reddit, and social media. I spend a lot of time writing each article, sometimes up to a week. My goal is to write evergreen articles that are still valid in 1-2 years' time.
It is a long tail approach. When someone new comes to the site, they have to chance to dig deep, see how you think, hopefully get to know you well enough, and in the best case scenario — you gain their trust and they end up supporting the project.
If you could go back and start over, what would you do differently?
Marketing over development. One of my biggest problems as a programmer is spending too much time on programming and developing new features. That's a mistake, because you can have the best product in the world and still fail if people don't know anything about you.
My approach over the years has changed from 80% dev - 20% marketing in the beginning to 80% marketing - 20% dev these days.
Marketing is a very difficult skill to master for a programmer, because it requires a different mentality altogether. With programming, you write the code and get instant feedback if it is working or not. Marketing, on the other hand is a long tail game and requires patience. Something might not be working today, but that does not mean it will not work in 2-6 months.
What's your advice for hackers looking to profit from their apps?
Don't spend too much time writing code. Marketing above all. Don't be afraid to do whatever it takes to get the word out. It's a noisy world, and you have to beat your own drum as loud as possible.
Where can we read more about you?
You can also leave a comment below, and I'll try to get back to you!