Lessons from Publishing Our Second Steam Game
Lesson from project on Steam after 2 months of development.
This year, I am launching Reima Project, which aims to release multiple small-scoped games within a year. Our second game is Deflect Boy, an arena shooter with deflection mechanic. You can get the game in the Steam here.
Within this post, I’ll share what we learned from developing Deflect Boy.
Background
Deflect Boy is the 2nd game in under Reima Project. After the release of Solar Serpent Squadron, we are still not satisfied with the amount of polish, juice, and fun elements, as stated in this post about it.
We want to make more visually polished game, but at the same time we don’t want to stray too far from Solar Serpent’s genre, so that we can take parts of that game into the next one. So the game should still be generally a shooter, with simple controls and a branching upgrade system ala roguelite.
Based on that, the idea of Deflect Boy is born.
The Game
The concept of Deflect Boy is simple : what if we build an arena shooter, but we cannot shoot by ourself, but deflecting enemy bullets instead. We got inspired from other games with unique bullet mechanics like Boomeraxe and Asterogues.
In order to make this game within 2 months, we want to limit the scope so that it’s manageable.
Initially we want to make this game endless with survivor-like loop, but we decided to make it level-based (15 in total), with closed-space arena. This is important because we want to pour our focus on the game’s feel and VFX.
Outside of the product itself, we also want to level up our marketing efforts. So this time we set the Steam page with polished key art (credits to our dear friends at Mankibo), we create a more polished looking screenshots, tailored the descriptions and localized them using machine translation.
We also setup our developer page and published some announcement prior to launch, and we published our store page around 45 days before launch, and we sent some keys to streamers and prepared a press kit here.
What went well
In respect to our development goal, more or less we achieved our target. We managed to “juice up” this game to the standards that we want. This is also proven when I shared the trailer publicly, we got positive feedback about the polish level.
The game’s concept is also quite appealing because of the unusual deflect mechanic. However I think we can still improve the concept even further.
As for the store page, using a polished key art also tremendously helping the appeal of this game.
Our screwups
In development of this game, we made some screwups along the way.
First, we didn’t do playtesting early. Initially the game doesn’t have auto aiming after player deflect the bullet. We assumed (wrongly) that this will make the game challenging and fun.
But when we do playtesting much later at almost 75% of dev time, we just realized that playtesters were having a difficult time hitting enemies with the deflected bullet. They have to focus at both deflecting and re-aiming the bullet at the same time, which splits the player focus.
Because the dev time is nearing release, we didn’t have enough time to make a proper fix, so we just “patch” the design by implementimg auto-aim. The result? Much better, but there are still other mechanic flaws that we found after release, that could be prevented if we playtest it earlier (and more frequently).
Our second screwups is a series of marketing mistakes. Firstly, we focus too much on development that we don’t have enough materials to do marketing variations. We just made the bare minimum (trailers, key art, some screenshots and GIFs). This made our marketing deliveries very limited, for example we cannot post many twitter posts.
Next, we are still unfamiliar with Steam’s dashboard. I didn’t know that we need to set the release date and displayed date at least 14 days in advance. Initially I set the date to 7 April and forget it until the real release date (which is 29 April). When we realized it, it was a little bit too late to change it via dashboard, so we need to contact Steam in order to change it.
Here’s the example setting in the Steam dashboard:
The same happens also with the price. We initially want to release it for free with DLC, so we paid the Steam direct fee with that setting. Later, we decided to release it as a paid game, and apparently, there is no way to change it in the dashboard! So we did what we did before, we contacted Steam to ask them to change it.
Lastly, we also screwed up our press release. We already prepared our press kit, but when creating the email body to be sent, we mistakenly set the release and embargo date TWO MONTHS EARLY than it should be! Our game is released in April, but in the email we said it’s on February 🫠. So check your email before you send it!
Lesson Learned
We want to release a series of small games so we can accumulate mistakes early and often, and learn as much as possible. Based on the positives and screwups that we made, here are the lessons that we learned.
In our case, we deliberately choose this game’s idea so that we don’t build it from scratch. We try to take as many elements as we can from the previous game. This is also the case with our next game, Dinner Defenders.
We MUST playtest early and often to capture as many user feedbacks as possible. This is crucial to avoid waste in development time and to shape the game better.
As we develop your game, we should record and capture some footage for marketing material. If we do it in bulk, it will take too much time and there will be no sense of “before/after” in the footage. Players love to see how the games are being shaped.
We also need to be mindful of our stated release date, as out marketing effort is built around it. The release date in Steam cannot be changed within 14 days of the date.
If we are still unsure if the game is paid/free, just set it to paid first. It’s easier to make paid games free than vice versa.
Lastly, we must really be careful for human errors when broadcasting press releases and keys, as mistakes made here is very critical. Have someone (or two, three even) look at your draft before publishing it.
Well, that’s all folks, hope you can also learn from our journey developing this game.
Thank you and see you again!
