It has been three months since the release of Cat Herder on Steam. In that time, I’ve put out two major updates and several smaller bugfixes, as well as started working on my next project.
Now though, I want to take a moment to reflect, to go over the project and its performance and see what lessons can be learned.
Before we begin: by its nature this post will focus heavily on mistakes and things I could do better. So to balance that, I should be clear that I’m generally happy with how this all has gone, and had a lot of fun along the way!
Let’s get started.
Puzzles: A Fundamental Conflict?
Here’s a question: is it possible to design a satisfying puzzle when the puzzle mechanics rely on random chance?
Some might call this a “Cursed Problem”, a fundamental conflict between plan-focused puzzling and the inherent instability of randomness. And I might be inclined to agree, which is why I spent so much time and effort trying to circumvent this issue when making Cat Herder.
When left to their own devices, the cats will wander randomly. However, using various toys, the player can control the cat’s behavior and direct them where they need to go. Every puzzle in the game can be completed in a deterministic way, there is always a concrete solution.
However, it’s also true that every now and then you might get lucky. Your approach to the puzzle might be completely incorrect, but if the RNG gods are on your side, you might get through anyways. This is a problem, because it teaches the player, incorrectly, that relying on luck is a valid strategy. Then, when they get stuck on later puzzles, their first instinct is to just bang their head against the wall waiting for the dice to come around, instead of reevaluating their approach.
I saw this happen repeatedly, first when my friends playtested the game and later when it was played by content creators. However, the issue was definitely way worse for the content creators, as seen when Sodapoppin, a Twitch streamer with over 8 million followers, ragequit the game after playing for just 20 minutes.
So why wasn’t it such an issue during playtesting? Well…
Playtesting vs Playtesting Effectively
Playtesting is always important, but how you go about playtesting is just as critical, especially for a puzzle game.
The game was still early in development when I started having my partner and close friends try it out, so I gave lots of hints and talked a lot about my goals for the design, and I think that’s fine.
However, after that I only tested the game a couple times, and only saw one of those tests in person. They didn’t seem to struggle too much, but that might have been because all my friends who had already played the game were there as well! It was valuable, but it wasn’t the fresh perspective that, in retrospect, I needed.
So, for the future, doing more playtesting and doing it better is key. Still, that’s not the whole issue. Even after seeing the problem play out across numerous videos, it took me a while to really understand why it was happening, and even longer to actually think of it as a bad thing. I mean, herding cats is supposed to be frustrating, right?
The Feedback Mindset
There’s something to be said about frustration as a feature, about the appeal of unconventional games and sticking to your vision, etc, etc. But there’s a difference between a player feeling frustrated because a game is challenging, and feeling frustrated because a game is poorly communicated.
That it took me so long to see that speaks to a deeper problem, that unless I am specifically in a “feedback” mindset, I am glacially slow to respond. If a player messages me requesting a feature, I’m on it. If I see a recurring issue during playtesting, I note it down. However, if I see multiple streamers miss critical information because the UI has a bunch of extra info that isn’t relevant yet, I apparently do nothing for a month and a half, before finally implementing a trivial fix.
I am just now, as I write this, realizing that I should really put in some loading screen hints between levels, so I can tell the player directly that none of the puzzle solutions require random chance. Why did this take me so long??
Of course, it’s hard to accept feedback objectively, even more so when the player in question isn’t having a great time. It can be easy to dismiss complaints, to say that they just don’t get it. But the correct response there is to ask why they don’t get it, and that’s a question I need to ask more often.
Marketing and Sales
Ok, switching gears now.
The game was more or less finished about a month before release, and I spent that time marketing aggressively, albeit clumsily. See below for a full breakdown of the various social medias / strategies I used.
My posts performed… fine. The game isn’t necessarily flashy and I’m not so sure about the color palette anymore, but it’s cute and silly and there are lots of places on the internet where you can talk about cats. However, I made the rookie mistake of not marketing at all during development, which was dumb. On the day before release, I only had 181 wishlists.
So how did I turn this weak start into a mediocre success? Well, if there’s one thing I did right in this whole process, it’s the opening scene of my trailer. All those cats rushing into the frame is super attention grabbing, and makes for an awesome thumbnail. I posted that video everywhere, and in a couple places I got lucky and it seriously took off. A good trailer is always important, and I highly recommend this GDC talk by Derek Lieu if you’re looking for advice on how to make one.
All that external traffic gave me enough of a boost that Steam itself also started helping. All told, about 53% of my traffic came from Steam. I apparently hit New and Trending, but I barely got anything from that, so it must not have been very high.
Here’s a look at my visits over time. You can see the big spike at release, a mystery spike on Nov 8th that I’m still confused by, and several spikes around the Steam Winter Sale. I timed a major update to coincide with the sale, which seemed to help.
As of writing this post, here are the numbers:
- Impressions on Steam: 952,251
- Steam Page Visits: 179,034
- Wishlists: 3,182
- Units Sold: 1,596
- Reviews: 30 (all positive?!)
All told, it’s less than I had hoped, but more than I probably had any right to expect. At this point, purchases have largely stalled, but I expect that I’ll see a couple hundred more during various sales.
Content Creators:
I manually reached out to a total of 376 content creators across Youtube and Twitch. Of those, 13 made a video, including some pretty big names like Sodapoppin and Ctop. Here is a more detailed breakdown:
I don’t really have a way to gauge the impact of these videos. It’s possible that the Nov 8th spike is due to Sodapoppin, but that livestream happened on Nov 6th, so the timeline doesn’t really make sense. Outside of that, there’s no obvious trends in the data that I can point to.
As a side note, manually researching and contacting all those creators was a massive pain, and I’m not sure it was worth it as opposed to just using something like Keymailer.
Reddit:
Reddit was definitely my biggest source of traffic, and that’s almost entirely due to this post on r/Cats. I still have no idea how it didn’t get taken down, but I’m eternally grateful.
I also messaged a bunch of users that had previously DMed me about the game, but ended up getting banned from Reddit for three days for “spamming,” so uh, don’t do that.
Twitter:
I didn’t really get Twitter at first, and maybe I still don’t. However, what’s become apparent to me is that, unless you get lucky with a viral post, growing a following on Twitter requires a fair bit of active engagement and effort.
That being said, I have made some great connections on there. In particular, I was contacted by a dev team that, completely by accident, put out a game called Cat Herders, with an “s”, soon after my game released. I thought it was pretty funny, and we both decided to just go with it.
Mastodon:
With all of Twitter’s… everything, lately, I thought I’d try this one out. Surprisingly, it’s actually become my most successful platform after Reddit, with the second most followers and store page visits.
I absolutely recommend checking it out, though like Twitter it requires active engagement, so keep that in mind.
Tumblr:
I posted here with basically zero expectations, and was surprised to actually get a fair amount of engagement. I don’t get tumblr at all, to be honest, but they like cats.
Tiktok / Instagram Reels / Youtube Shorts:
The nifty thing about these platforms is if you make a video for one, you’ve already made a video for the other two. That being said, following the various trends and editing the videos takes a lot of time, and even when they do well people aren’t likely to visit your store page. I wouldn’t personally recommend this one.
Final Thoughts
At the end of the day, I think this whole thing went “fine”. It wasn’t a huge success, it wasn’t a complete flop. The game has issues, yes, but it has a lot of good points too, and I’m proud of them. I’m also proud of myself for making it all the way through, for developing and marketing and releasing a game. I’ve learned so much, and I plan to keep improving.
Cheers!
-Harrison
Switchback Studio 