On November 2023 I quit my cushy Data Scientist job to make the roguelike autobattler Gods vs Horrrors (GvH). On May 5th 2025 I released it and in its first month and a half it’s sold over 7k copies and made me a bit more than 25k$. In this post I’ll talk about my journey and learnings during the process.
Am I satisfied with the result?
I’m very happy with the game we made. I still enjoy playing it after dedicating 1,5 years of full-time effort into it, which is a lot to say.
I’m less happy with its commercial performance. I think it’s received a lot less attention than it deserves, and that many roguelike deckbuilder and Hearthstone Battlegrounds (HS BGs) players would enjoy it. For comparisson, I feel like The Ouroboros King (my previous game) got more attention than it deserved.
I’ve spent about 120k$ in this project, ~30k$ in direct costs (marketing, illustrations, localization, software, …) and ~90k$ in opportunity cost (I worked on it for a year and a half and my previous full time salary was around 60k$/year). I’m not sure I’ll make that money back. I’ll be OK financially since TOK was a lot more profitable, but I hope to do better on my next game.
The review situation
As of July 17th 2025 Gods vs Horrors is sitting at 77% positive reviews on Steam, which I find quite disappointing.
The score is so low due to a 70% ratio on Simplified Chinese reviews, which amount to 64% of total reviews (while only being 45% of players and 27% of revenue). I attribute the low score to a mix of different genre expectations, too wide marketing on China (thus getting players from other genres), and bad localization.
The average score for non-Simplified Chinese reviews is 94%, about which I’m a lot happier.
Why did Gods vs Horrors not get as much attention as I would’ve liked?
- The game came late, and it was already late when I started making it. It’s been 6 years since Dota Autochess started the autobattler craze and 8 years since Slay the Spire was released. With this time autobattlers and deckbuilders have evolved a lot, and Gods vs Horrors doesn’t feel like it brings anything new. It’s genre could’ve been a strong hook 4-6 years ago, but not anymore.
- It’s a relatively hard game to get into. To fully enjoy the game, you need to know most cards in the card pool, so you can decide the direction of your build based on what cards you can get in the future, not just what you’re shown now. Early play-throughs involve lots of reading and that can be a turn off if you aren’t already invested. For contrast, HS BGs had the advantage of using cards from the original HS, so players were already familiar with them.
- Autobattlers are a genre full of free alternatives and relatively low-priced games. By contrast, deckbuilders tend to command higher prices.
- I feel I’ve been unlucky with streamer coverage. The only big streamers to cover the game on release were Olexa and SIFD, and I’m very grateful to both of them. Retromation would’ve probably covered it if he wasn’t mostly playing the Bazaar. And I think lots of other streamers like Northernlion, Kripp, and all the HS BGs streamers could have enjoyed it. GvH youtube videos amount to ~200k views, while TOK (which got lucky with Aliensrock coverage) got over 2M.
- The game’s aesthetics and juice aren’t competitive at the level of success that I’m aspiring. I was afraid to make a big investment into art so I told Isaac (the game’s illustrator) to make stylized illustrations. If I could go back in time, I’d tell him to spend more time (and charge me more) for the illustrations. Additionally, adding VFX, SFX and game juice is not my strongest skill, and the game could be improved in that desparment.
How I’ll do better next time
Game success comes down to 2 things:
- Appeal: how likely Steam customers are to buy your game after looking at its Steam page. After all, that’s what most players look at to decide if they’re going to buy your game.
- Gameplay quality: you also have to deliver on the gameplay or you’ll get terrible reviews which tank your sales and drive up returns.
My previous games have lacked in mass appeal (moderate-low sales), while being fine in terms of quality (high review scores).
I feel confident I can deliver in gameplay quality if I pick a project that isn’t out of my scope. So I need to get better at picking the right project.
In the future I want to iterate faster, making early prototypes and trying to get attention to them. This should help me find winning projects, and avoid wasting time on games nobody likes.
The development journey
Developing as a solo developer has been a lot more stressful than doing so as a side project. It’s the difference between the game being a nice bonus, versus your main source of income.
This increases pressure. You want all your announcements to hit big, which makes the previous days a lot more stressful.
I’m lucky I have some savings and basically used my previous game’s revenue to finance this development. I can’t imagine doing this otherwise.
Development time
The project would’ve been more profitable if I hadn’t wasted some time and resources.
Here are some thing I should’ve avoided:
- Doing the UI myself. I did the UI for TOK and thought I could do it for GvH as well. It turned out to be a bad idea, as I wanted the game to look a lot better than my previous one. I spent too much time on it and ended up having Isaac (the game’s illustrator) redo it. Going forward I should probably not make any graphical assets myself, and work with professionals.
- Hiring a marketing person. Took too much time onboarding and managing her, and the results were almost non-existant. We didn’t get traction on any social media platform. Half the problem was that the game was hard to market and the other half was hiring someone inexperienced.
- Trying to make a side business. Since I didn’t have enough work for the marketing person I’d hired I tried to make a service out of the streamer data pipeline I’d developed for personal use. It never took off and I wasted about a month on it.
Marketing
The only stuff that reliably moved the needle for GvH were content creators, Steam Next Fest and whatever my Chinese pub is doing in China.
I had a marketing person regularly posting on social media on my behalf and we never got traction. I understand it may be worth it for more broadly appealing games, but not in my case. I agree with Chris Zukowski’s stance on this: 90% of marketing is picking a marketable game.
The only promotion that is so cheap everyone should do is emailing streamers, however that doesn’t guarantee any results. It’s like almost free lottery tickets.
Since I wasn’t getting much coverage, I also tried to sponsor some streamers. I spent a ton of time researching which channels were truly suitable to my game, and making estimates on how much I could afford to pay them.
I made sponsorship offers to 45 creators based on sharing 50% of the value I expected to get from the streams. Offers ranged from 50-3000$. I was declined by all except the 2 streamers who had previously played the game (and would have probably played the game on release anyway). Another 4 replied saying their rate was a lot higher.
I knew I was offering below market pay, but I tried to compensate with flexibility and convenience (pay upfront, no revision, letting them do multiple sessions if they like the game). It didn’t work.
It was a huge time sink, and wouldn’t recommend anyone else do the same.
Conclusion
I consider Gods vs Horrors to be a moderate success overall. It allows me to keep making games but hasn’t fulfilled my expectations in terms of commercial success. In the future I need to be more efficient and focus on making more appealing games.