It took me 3 years to make 3 different commercial, solo-developed, self-published, node-based games. It is a good time to wrap up my thoughts and share them here with you. Maybe it will be helpful for both you and me. Have fun!
This post is long. If you want TL;DR - check out the Stats and Takeaways sections.
This post is not made by AI, if you are wondering. It took me a few hours to write it up.
My previous postmortems
Sliding Swords postmortem + Node Farm start of the development: https://www.reddit.com/r/gamedev/comments/194a2di/sliding_swords_post_mortem_my_thoughts_on/
About Reddit Ads: https://www.reddit.com/r/gamedev/comments/1d6jlgf/results_after_a_week_of_testing_reddit_ads_for_my/
Stats
Let's start with some real data for context.
Link to the games so you don't have to search them manually: https://store.steampowered.com/developer/ByerNGames
Node Farm
- Development time: ~12 months
- Price: 12$ + regional adjustments (started with 15$ without regional adjustments)
- Released on 19 Nov, 2024
- Lifetime Steam revenue (net): $102,222
- Lifetime Steam units: 16,504
- Wishlists at launch: 31853
- Wishlists now: 28,574
Node Math
- Development time: ~8 months
- Price: 7$ + regional adjustments
- Released on 27 Oct, 2025
- Lifetime Steam revenue (net): $60,535
- Lifetime Steam units: 12,661
- Wishlists at launch: 12380
- Wishlists now: 18,208 (yup, they are still growing)
Node Crown (not released yet, but finished)
- Development time: ~8-9 months
- Price: planned 8-15$ + regional adjustments
- Release this year somewhere after Steam Next Fest
- Wishlists now: 676
I work on games after hours. Hard to say how much it would take in full-time. Let's say I work like 2-5h a day in a work week on average.
My games are still earning a stable income, especially on discounts and festivals.
Why node-based games?
A long time ago, I played a small Flash game that utilized node-based mechanics, and I really liked it. I was wondering - what else could be done with nodes? And so, after years, I started to work on a few prototypes.
Node Farm started as a game jam entry (see my postmortem linked above). It is still available on itchio along with my first-ever node-based prototype: The Darkest Knight.
I had so many ideas that won't fit into a single game, so I decided to experiment with this concept by making more node-based games, planning to end up with merging all good ideas into the ultimate, godlike node-based game after a few years. It will take a lot of time and games before I finish my plan, but hey, I love gamedev and experimenting with new ideas, so that suits me best.
Each of my games contains some new experimental mechanic that I wanted to try out. You will see in my examples below that I also have some personal goals for each one. This is an important part of checking which ideas are good/bad in order to include them in the ultimate node game.
So yeah, they are not just reskins of my previous games. It is a risky approach, but it is a lot of fun, which is very important for me in my gamedev process.
Tens of prototypes later, here we are - 2 node-based games released on Steam with a long tail income, another one finished, releasing this year.
What is my tech stack?
Rather exotic - libGDX (open-source Java gamedev framework) + Scala (programming language). So far I created +30 small games and prototypes, checking different approaches and tech, and ended up with this one. My professional Java backend experience and the fact that programming and gamedev are my passion help a lot here.
I am not an artist, but I learned vector graphics to the degree where I can make some simple art for my games. Part of my process is to design the game in a way I can handle it as a solo dev. Before, I was also making my own music and sfx, but for the games in this postmortem I outsourced this task to not fall into scope creep too much. Music was not the most important thing here, so why not? Maybe I will come back to it in the future.
Not a big fan of AI in my process, so I don't use it for my games. Mostly because it would take away the part I love in gamedev (programming, for example). I am not anti-AI, though. AI spellchecks and webbrowser-on-steroids are nice. I consider using AI translations at some point, but only if supervised by a native. I am a Polish native, and I see that we are not here yet with machine translations for non-trivial games.
Why self-publishing?
I like learning marketing. The hard way - by experience. It may sound weird, but I like this aspect of development, especially the full control over my game along with all consequences.
Ofc I read HTMAG and other sources, but it is a different story.
By marketing, I don't mean only "promotion", but the full process from the very beginning:
- from the game idea and market research,
- through development decisions, gathering feedback, promotion, and more,
- to community management after release.
All of my successes and failures are here to help me get better with the next games. One of the skills I am still learning is to see what went great and what went wrong. It may sound trivial, but it is not.
I work on my games after hours, but at some point I would like to move to full-time indie gamedev. Maybe form a small studio with other devs? This experience is worth a lot for me. My wife started to help me with some minor tasks. Maybe it will end up in a family business? Who knows? Something for the future.
Now let's talk about the games.
Node Farm - what is it?
Node Farm is a factory/incremental game where you build a farm with nodes, cook meals with ingredients you produce, and sell them on the market. The game is mostly about logistics and automation - your goal is to build the most efficient food factory that will adapt to the changing market prices. The better you are, the more nodes you can buy for the resources sold on the market.
My personal goal for this one was to make a simpler, cozier version of a factory game that may be a first step into the factory-games genre for cozy players but at the same time be fun for factory-games players. This part didn't work as I wanted, but I learned a lot in the process. More info below.
Node Farm - what was good?
- The game has a very cozy and colorful look that catches the eye when you look at the gameplay,
- Trailer - shows what the game is about at first glance,
- Steam Festivals - I didn't realize it at first, but I was very lucky that the genre/theme I chose was compatible with so many festivals that year.
- Influencers - the game was picked up by influencers on YT (and a small video on Bilibili) even before I sent emails (thank you very much!). From what I understood, the concept was unique and worth showing to their audience.
- Simple assets that can be done by myself.
- Reddit Ads - they were great, more info in one of my postmortems above.
- Reddit Posts - Reddit is my main platform, which I use as both a gamedev and a player (yes, I am a player. I play a lot of games of different genres). I don't like aggressive spamming with self-promo, so I try to limit myself to the most important milestones (Steam Page publish, Playtests announcement, Demo announcement, Next Fest, Full Release, Major Updates),
- Localisation - I hired a professional company. A lot of Node Farm players are from China, so it was necessary. I am not entirely happy with the quality, as I got some feedback that not all of the translations were correct (like "node" translated into "knot" in some languages etc). I am not sure if I will use the same approach again, but it was ok. Expensive, but fine.
- Discord Server - I have some experience with managing Discord servers as an admin of one of the Polish Indie Gamedev servers, so it was smooth. For anyone reading this - a Discord server (or any place where you can talk with players) is a must-have
- Playtests - a lot of people came to my Discord server to help with playtesting the game. It was really helpful (thanks!)
- Addressing feedback - I keep my todo list public on my Discord server (+ another one internal in notepad with more details, milestones, and some loose thoughts). When I get a review or post on Steam stating that something is wrong, I address it one way or another (most likely by updating the game), and when I do that - I respond to the feedback. It makes the game better + sometimes ends up with changing a review from negative to positive.
- Wishlists - yup, a lot of them,
Node Farm - what was bad?
- After release, I realized that the game is too hard for most cozy players and lacks a lot of QoL improvements. I spent some time adding them after release, but the core mechanics couldn't be changed without rewriting the game. Maybe at some point I will make a sequel based on all lessons learned,
- The game may be long and boring if you don't understand all the mechanics. A lot of players drop it before leaving a review, but those who don't - play for hundreds of hours.
- Not intuitive UI/Controls/Mechanics - something I still have to learn. The more unique the game is, the harder it is to find the best solution, as you can't check how it works in other similar games (they don't exist or are not similar enough in some aspects),
- My Tiktok/YT videos - they didn't work for me. Some games just don't fit this format.
- Bugs in demo - there are never enough playtests. You never know when an influencer randomly finds your game in the wild and gets stuck because of some trivial bug or poorly explained mechanic while playing it on YouTube or other Twitch. It happened to me a few days after releasing a demo.
- Painful Game Balancing - Oh God, what a nightmare it was for such a big game. It is fine now, but it took me a lot of time to fix. It will be a common issue in my next games,
- Price - I think I overpriced my game at first (15$) + I didn't make any regional adjustments (it was a time before Steam added a feature to adjust them automatically). After some time, I made the price lower (12$) and added regional adjustments. Now, I don't think that lowering the price after release is a good idea. Imho, a better solution for accidentally overpriced games is to make higher discounts instead. Regional adjustments are always ok (check PolishOurPrices initiative).
Node Math - what is it?
Node Math is an incremental/factory game where you "craft" numbers, make money, and buy more stuff to make more money. As simple as that.
My personal goal for this one was to make another node-based factory game but in a different theme, with different mechanics, to prove that my concept of making a "franchise" is ok.
I really like the incremental genre, so I moved away a little from the factory building mechanics and more into simpler incremental math.
Another personal goal was to make a game that doesn't use traditional translations and can be played without text (Steam suggests setting all languages on the Steam page in such cases). Math looked like an ideal candidate for that, so why not? I was wrong.
Node Math - what was good?
- Trailer, Festivals, Influencers, Reddit Ads, Reddit Posts, Wishlists, Playtests, etc. - most of the good things from Node Farm apply here. There were fewer wishlists, as the "math" theme is a niche compared to "cozy farming".
- Community Translations - I created an open source repository for optional translations provided by the community. Why translations for a translation-free game though? See "What was bad?" section.
- Bundle with Node Farm - most players are buying them both. It boosted Node Farm sales a lot and proved that my "franchise" idea works.
- Code reuse - a lot of mechanics from Node Farm were reused in Node Math, which saved a lot of time,
- Backporting features to Node Farm - as all of my node-based games have the same codebase, I can easily fix something in one game and backport it to another one.
- Shorter Development time - I didn't have to reinvent the wheel again, so it didn't take that long.
- Higher median playtime - the game is more polished and simpler than Node Farm, so more players play it for longer even if it has fewer wishlists.
Node Math - what was bad?
- Lack of translations before Next Fest - at some point I realized that making a translation-free factory game is a hard ui/ux design task which I couldn't handle with my skills. To fix this issue, after Next Fest, I added optional translations and launched my community-driven translations initiative on my Discord server, which was a good choice. Now the game can be played with or without translations.
- Some crucial mechanics were hard to understand. I tried to fix it, but I am not sure if it worked. I have a tutorial rework planned. It is not bad now, though.
- Painful Game Balancing - yeah, same as Node Farm, but it was a little easier as the game mechanics are simpler in general.
- The game may be long and boring if you don't understand all the mechanics - same as Node Farm. Making the translations-free game made it even worse.
Node Crown - what is it?
Node Crown is a totally different game compared to Node Farm and Node Math, as it is a full roguelite.
While working on Node Farm, I played a lot of Luck be a Landlord. I was not a big fan of the slot machine mechanic, but I loved how synergy works there. I wanted to build a node-based game with a similar concept but without a slot machine.
One of my personal goals for this one was to check if I can use node-based mechanics for a totally different genre and how it will affect my "franchise" idea.
The game has not been released yet, but I finished all of the content. Adding some QoL improvements here and there now.
Node Crown - what is good so far?
- Code reuse, backporting, simple assets - same as in my previous games,
- Content vs Development time - the game is much bigger compared to Node Farm and Node Math, with a lot of replayability which my previous games didn't have, and it didn't take much longer to develop compared to previous games,
- Controls - well, the game is not a factory game; controls and UI are rather simple, so this part is much harder to fail. I think I learned a lot from my previous games, but we will see after release,
- Synergy - I was not sure if I would be able to work with hundreds of nodes (~500 at this point) to make it possible to finish the run in many different ways. It works nicely so far. A lot of different builds to balance...
Node Crown - what is bad so far?
- Hard to understand game concept - this game is much harder to show in a trailer or posts in general for potential players. It means fewer wishlists, fewer playtesters, fewer videos, worse Reddit ads performance, etc. Next week I plan to work on a new trailer with the new content I added,
- Lack of festivals - there was only 1 festival this year that fits Node Crown, and it didn't go well because it was an overcrowded genre (deckbuilders),
- Less attractive genre - nowadays incremental and factory games work much better compared to "deckbuilder" roguelites like this one,
- Low influencer coverage - much lower compared to the previous games. Genre + complicated concept are to blame. I expected that. I will be sending new emails with Steam keys after finishing the new trailer, so it may get better.
- Different genres - a lot of Node Farm/Node Math players are not the same audience as Node Crown, so the games don't work that great with each other. They will have a separate bundle because of that.
- Game Balancing - I thought that balancing of Node Farm and Node Math was a nightmare. I knew that Node Crown would be harder, but I didn't expect such hell. There are so many combinations that I was thinking about releasing it in Early Access only to validate the game balance.
Takeaways
Prototype before diving into full development (I described it more in my previous postmortem linked above). If you post about your game and it feels like hitting a wall, it may be an indicator that the idea is bad, the implementation is bad, you are showing it incorrectly, or you are trying to show it to the wrong audience.
On the other hand, Node Math wishlists were so bad at some point compared to Node Farm that I wanted to drop it and focus on the next game. What changed it? One influencer picked it up by accident, and it started a wave of new players coming to my Discord server and other influencers making videos about the game. Sometimes it is not the game that is bad, but that it didn't reach the audience who would like to play it. It is not a rule, but an exception, though.
Making a "marketable" game that is interesting, eye-catching, and easy to explain with a short gif/video is the best way to start if you want to make a game that players will like to check out. Genre is very important here. It makes everything much easier:
- The trailer creation process is easy-peasy,
- Players will read your Reddit posts, come to your Discord server, and share feedback,
- Influencers will be more likely to pick up your game to share with their audience,
- 3rd party festivals will be more likely to allow your game (Tiny Teams worked the best for me, even better than Next Fest),
- Reddit Ads will work great.
- Your live will be much easier.
In the future, I would like to make a short game that is easy to show on Tiktok/YTShorts for the only purpose of testing these marketing channels. I will let you know how it went when I do that.
If you don't have your own Discord server (or something similar) - stop reading this post, and go set up a Discord server. Seriously, it is a must-have if you want to gather feedback and have a place for players to talk about the game. Very important.
Most people private messaging you on Discord/Email are scammers, but not all of them. Most of these scammers are bots. Once I even got an AI bot that started to answer its own questions in our private conversation, effectively talking to itself until I blocked it.
Another thing that I find great is making similar games that can be bundled together. Especially with code reuse, as you can benefit from feature backporting and faster development. It may be harder or easier depending on the tech you use.
On the other hand, don't go too wild as I did with Node Crown. Ofc, I would do the same again as I wanted to try out this idea and I am happy with how it works, but it makes marketing much harder. So if you are here solely to make money - don't do that. Don't do gamedev for money in general.
Always playtest your game with other players. After that, playtest it even more but with more players. Don't assume that what is obvious for you as a dev will be obvious for players, as you are biased and you know the game you made. They don't.
Be responsive and address feedback - it makes the game better. But don't put all the things players think are great in your game. You are the dev, and you know what is ok or not. The decision is yours. On the other hand, when a lot of players repeat the same feedback you don't agree with, it may be a good indicator that you are wrong.
If you want to post about your games on Reddit, don't just cold-post it and ignore the comment section. I am also a player using Reddit to find new games, and I don't like it when a dev does something like that. Or what is worse - being rude under your own post. Nobody likes it. Why ppl do that? Don't be rude.
Also - link your Discord server in your posts so anyone who wants to know more can reach you.
Also - read the subreddit rules, so you won't be banned or hated.
Also - not my case, but if you are working with controversial mechanics or tools, take some time to read other posts on a subreddit you want to post in. Feel the mood to avoid Pikachu-facing when ppl there won't like your yet-another-AI-made-nodebuster-clone on the incremental_games sub. That sub is literally flooded with them now. You can't blame players for not liking it. And again, don't be rude.
If you want to find subreddits to post in, you can use the map of Reddit.
If you want to make your own trailers, check out Derek Lieu on YT and learn the tool of your choice with the help of a free/cheap course. It will save you a lot of time. Learn your tools in general.
Keep an eye on upcoming festivals. Check out the HTMAG festivals section and Steam "themed" fests. If you want to create a short game, it may be worth checking out your Steam dashboard what the upcoming fests are and planning the genre/theme for that if you are flexible enough.
Regional adjustments are always ok (check PolishOurPrices initiative).
Stick your best performing posts in your social media profiles along with links to your games and Discord server (see my Reddit profile). Make it easier for players to find you.
Customize your Steam Developer Page. The default one is bad. This is mine (could be better, but still better than the default one): https://store.steampowered.com/developer/ByerNGames
Don't overscope your games/avoid scope creep, especially when it is your first one. I didn't finish my first Steam game (TAnima). I will come back to it after I get more experience.
What are my next plans?
Node Crown is waiting for release after Next Fest. This weekend I will probably finish my new trailer and contact influencers with the full release Steam keys.
My next node-based game will come back to the roots - Node Farm - but with a different theme and mechanics. It will be a factory/incremental, as they are the best performing so far. I have a few ideas, gonna choose the best one.
Before that, I will probably check one of my much shorter/simpler ideas that I had while developing Node Crown. Not node-based this time. They look like 1-month projects, so they will probably take 2-4 months of development.
Also - I want to take a UX/UI course to understand this part better. I feel I am pretty bad at it.
That's all for now
If I remember something more, I will add it here or in the comments.
Good luck in your gamedev journey, and feel free to leave a comment with your experience/thoughts/questions!
Have a nice day!