How I made side income from Jetpack Compose: My journey

For 10+ years I had been posting Android tips, opinions, and code snippets on X (previously Twitter). Nothing polished, just honest takes from my day-to-day work. I normally tried to have a critical mindset and expose my honest perspective on things when I found it valuable. And a little bit of raging from time to time as well!

I also spoke at national and international conferences like KotlinConf, Mobilization in Poland, and many others. Started small, then went for bigger events. That helped me reach new audiences, connect with devs around the world, and position myself as someone others could rely on for valuable Android content.

I didn't have a product in mind. I just liked to share what I was learning, both online and on stage, and over the years I grew a good sense on identifying what topics people were most interested in. But these things compound. Consistency + visibility builds trust. By the time Jetpack Compose came around, I already had a public image as someone people turned to for Android knowledge.

That was the foundation for everything that followed.

Then came the announcement of Compose alpha 🙀

I'm sure I was not the only one, but I could immediately see how big it was going to become. Coming from my previous experience building sites in React, the parallels were obvious: declarative UI, smart diffing, state-driven rendering, lifecycle-aware effects... It was the same mental model all over again! but this time adapted to Android. And the fact that part of the original Compose team had worked in React before made it even more obvious.

It was clear to me: Compose wasn't just another library. It was the immediate future of Android UI, and every company would eventually migrate, inevitably. The only question was how soon.

That's when it hit me: if this was going to redefine Android UI development, then there would be demand for people who really understood it deeply.

On top of this, I also knew that diving into the internals of the View system had been a huge leap forward for me in the past. I knew that offering the same perspective for Jetpack Compose would help others as well.

Only a few days after the first Compose alpha release, I had already subconsciously made the decision to write Jetpack Compose Internals. And I wanted it to go deeper than just "how to use Compose." I wanted to explain how it worked under the hood, and help people to use it correctly and efficiently.

But timing was everything. I made it my goal to release the book on the exact day Compose 1.0 stable dropped. If Compose was going to make a splash in the industry, I wanted my book riding that wave. But I had a significant problem: The book was just barely started and there wasn't enough time to finish it.

To make it possible, I chose Leanpub. This platform would let me prelaunch the book with only part of it written, then keep delivering updates. That would allow me to hit the deadline without compromising the quality of the content.

Finally, I decided to not work with a publisher. Why give away a big cut when I could do it all myself? Writing, editing, promotion... yes, it was more work, but the audicence was already there, and it also meant full control for me. Publishers take a huge cut from authors today, to the point of making the deal very unfair.

The strategy worked. On Compose 1.0 launch day, my book was out too. That timing alone positioned me as "the Compose guy" overnight. I became the go-to reference for anyone looking to understand Compose internals.

Numbers were great for such a niche book: First month revenue was around $15k 🎉.

After that, sales dropped gradually every month, and ended up stabilizing around $2k per month for quite a while. This was a huge motivation boost for me to keep writing what I wanted to be the best Compose book in the market.

Even years after that, it still brings in $500–$1k per month, depending on how much I promote it.

This is how normally tech books behave. You can be lucky to get a viral boost at launch, especially if you have a good strategy and timming. Maybe some months after if too. But after that, it's down to consistent promotion and SEO. If you stop talking about it, people simply stop buying. That is something that I have learned during this journey.

Fast forward to today: Compose is the standard for building Android UI. Companies across the world are migrating. Almost no teams out there are sticking with the View system anymore.

That decision I made back in the alpha days proved to be the right call. And that's the thing with tech trends: if you catch them early, you can ride them for years.

One myth to clear up: no product "sells itself". There might be some exceptions, but in general that is really hard.

Yes, I had great timing. Yes, the topic was hot. But without me actively promoting the book on X, on my newsletter, and in other social networks, it would have faded out much faster.

Promotion is a continuous job. It doesn't feel glamorous, can be boring, but it's the difference between something dying after launch and something that keeps paying back.

The book was a success, but I knew not everyone learns from books. I also had some good skills giving technical courses that I acquired when working as a trainer for another company, so I decided to build a course and expand on all the topics in depth. My mindset on this was to not create a course that would barely scratch the surface, or just recap the book, but to go even deeper and cover all the topics in much detail. I really wanted it to be useful and worth for people to pay for.

I also decided to make it a cohort in the first place: I would record all the content, provide a support channel for doubts and interesting conversations, and deliver it over several weeks in a structured way. That would allow me to expand the content so much and explain all the topics as clearly as possible.

I built the ✨ Jetpack Compose and internals cohort-based course ✨. And it absolutely blew up. It was an absolute madness that I can't even believe to this day, honestly. Total revenue ended up at around $183k.

The cohort approach had its advantages:

• I could charge a big amount per attendee: $300–$500. I came up with that price with two things in mind: the continuously raising number of companies that were adopting Compose in the market, and my own exposure and positioning as an expert/authoritiy on the topic. I leveraged that, and I firmly believe that without those two things this price would have obviously been way too high.
• Private companies could book it for entire teams, even ask for private cohorts just for them.
• People would prioritize Compose for their company learning budgets, since mostly every Android dev wanted to learn it.
• Each cohort could have around 50 attendees, bringing in ~$20k per run. The recorded format + support chat made this possible and realistic. That numnber of attendees would be very hard to handle with a live course.
• I could run a cohort every 3–4 months to fill the spots. It was intense, but the live + recorded format created a lot of engagement.
• I could make the course way longer than normal.

After about three years, filling cohorts started to get a bit harder. Not because the content was less relevant, but probably because most of the people from my audience that could be interested in it had already enrolled. There is a point where this starts happening, and you need to chose between promoting it even more heavily by finding new promotion strategies, or switching to a more sustainable model.

This time I decided to try something different, and shifted to a self-paced course. Lower price, open all the time, wider audience. It still includes most of the benefits, inclduing the support chat and full access to the Effective Android Discord community, but it removes all the waiting. And you can enroll and take it whenever you want.

And honestly, after years of running live cohorts, I wanted a model that was more sustainable for me too. With two kids at home, it was getting harder to find time to organize the cohorts and all the logistics around it.

Important detail here: I did all of this while working full time. First at startups, then at bigger companies like X and Disney. I would put time every night and weekends to work on it. Luckily I was very efficient creating courses, so the effort was reasonablly manageable.

I never quit my job to do this. And I don't recommend anyone does until they have stable, sustainable income. Side projects take years to compound. A salary pays the bills in the meantime. Be bold about this move, and do it the right way.

That is probably the real deal:

• Speaking at events gave me visibility.
• Years of posting free content built trust.
• That audience fueled the book launch.
• The book reinforced my authority and helped sell the course. (Doubling down on an already validated topic).
• The course built even more trust and reach.
• My X account grew to 15k followers, giving me a bigger audience for future projects. So it keeps compounding.

It all stacks. One step amplifies the next 🔉

And one more thing: don't keep your community on a single platform. Twitter can change the rules any day. I've been diversifying across different social networks, newsletter, and other communities. Always build more than one channel.

Note that I didn't build a million-dollar app. I didn't freelance every evening. I shared what I knew, showed up consistently online and on stage, picked one topic at the right time, and doubled down on it. That turned into a book, then a course, and side income that's still paying back today.

If you're an Android dev thinking about side income, here's my advice:

• Share what you know.
• Double down when you see momentum.
• Think long term.

It's slower than freelancing, but the compounding effect makes it much more powerful and keeps building up over time, so it can only improve.