Hey folks,
It’s been a little over four months since Bubbles was released on Steam; it’s now out on three platforms and that’s given me enough time to assess what I think went well with the game and what didn’t.
With thanks to Jojoo for this art! Check out the rest of her work here!
What Was Bubbles’ Goal?
Before we dive into the breakdown, it’s important to understand what I was hoping to achieve by developing Bubbles. It was - first and foremost - a learning project for me. I was learning a lot about how to program a game on the job and I’d never released a game on Steam before. I was also managing outsource myself and performing daily, more menial business stuff.
That said, it was important to me that I made a good game - that is to say a critically well-received game. With no-one else to answer to, this meant I’m entirely responsible for ensuring whether the game is fun or not.
It was - crucially - not the game I had picked to make millions of dollars. Platformers are kind of notorious for being hard to stand out on any of the big storefronts unless you do something really, REALLY special; I certainly don’t have the budget or time to compete with games like Celeste!
I’m very glad that I picked this route, as I’d read many times that you should never make your first project your dream game as you will inevitably make mistakes.
What Went Well!
- The game is - at the time of writing, at least - critically well-regarded. The game has received some fantastic write-ups from the gaming press, including Kotaku UK, TechRadar, 148Apps, Pocketgamer and Keengamer. It also has good reviews on Steam and other platforms and a lot of YouTube and Twitch content creators who play the game seem to be enjoying themselves! I’m really happy that I managed to make a fun game above all else.
- The prototyping went well. I managed to come up with a fun game pretty quickly and iterate on mechanics that worked well (the different bubble powerups) and strip out those that didn’t (the Lemmings-esque puzzle/platforming aspect). Getting this right early paid massive dividends later on as it made level design so much easier. I’ve previously worked on projects where these elements weren’t quite right at prototyping and you have to resolve these gameplay elements around content that has already been built; suffice to say, this approach costs you time, money and sanity.
- The project was a good fit for my goals - I got to develop a decent understanding of coding, learn about what was involved in publishing a game to Steam (and what to avoid) and found out just how much other stuff is involved around making games that isn’t er, actually making games - e.g. doing tax returns, contacting people, going to events, social media management etc.
- Music was an amazingly good fit. Although I initially wanted a more pure chiptune soundtrack to fit with the rest of the game, I trusted Dan’s judgement and it absolutely paid off. Trust your experts.
- The game is reasonably polished. Barring a handful of weird, relatively minor issues with the cat customisation options, the game seemed to launch without any horrendous bugs that caused crashing, lost saves or progress issues.
- The game has decent level design (with some exceptions) - players seemed to by and large get what I wanted them to do and I think there are some levels that have a cool, fun and creative theme to them. I wish I’d cut some of the fat off a few levels and done more of them - perhaps having 30 levels per world, but have them be about a third to half of their current length.
- The Boosts feature got a lot of love. I’m absolutely delighted with this as I’ve heard a lot of players say that they enjoyed the game when they might not have been able to do so. Accessibility was always a very important part of the game’s design and development to me and I’m thrilled that I’ve been successful here. I actually only wish I’d expanded this further with features like a rewind function, on-demand savestates or the ability to reduce the game speed. This will continue to be a core pillar for each project I work on.
What Didn’t Go So Well
- Level design went slower than expected. I'd budgeted about a day per level, but that excluded difficulty balancing which happened later on and doing an art pass. These art passes took way longer than anticipated and didn’t factor into prototyping. I should have done at least one level with a full art pass and then re-evaluated my estimates here.
- The game is too difficult in general. At the time of writing, I think something like 3-5% of players will see the end of the game in some form or other. This is supported by the game’s achievements breakdown (though it’s not quite the full story as it doesn’t include players who use boosts to complete certain worlds):
I wanted to create a game that would get challenging towards the end, but it seems I overdid this with the World 3 levels in particular being a big, big spike in difficulty - only 7% of players finished it!
This partly came about as I ended up committing most levels to muscle memory whilst making them. Unfortunately, this had the side effect of me being completely unable to judge difficulty. Luckily, I had some friends who helped me test the game, but I probably should have seen more alarm bells when people weren’t progressing much past World 3.
I was able to tone down some of the more punishing aspects of some levels in updates to the game, but I also didn’t want to make sweeping structural changes to levels when the game has a time attack component to it.
- This one is hopefully not visible to players, but my UI code is appalling. I’m hardly the best programmer as it is, but every time I wanted to change something in the UI, it was a serious headache and this probably added at least a week’s worth of work. I should have refactored this rather than trying and hodgepodge a lot of clunky fixes together.
- As mentioned towards the top of this breakdown; from the point of view of picking a game to make an untold fortune, this wasn’t the game to pick. I think beyond being incredibly hard to stand out as a platformer on Steam (with incredibly stiff competition from the likes of Celeste and Super Meat Boy), the game was positioned somewhat awkwardly.
Specifically, I suspect that more hardcore gamers would see the adorable art style and single button controls and think the game would be too simplistic for them. More casual gamers would see the terrifying World X levels and think the game is way too hard for them.
Whilst this is validation of how hard World X looks, I think this is indicative of the fact some people might have been put off by the game!
My trailers attempted to appeal to both, but perhaps it would have been better to release two trailers focusing on different aspects of the game.
My marketing push didn’t go particularly well. I think I started it too late (I started pushing the game about a month before launch and sent out very few pre-release versions) and didn’t spend enough time looking into content creators to share the game with. I also think they needed a bigger lead time on making videos.
I also don’t think I did a great job with the emails I sent out - whilst I tried to cut down on the text, I think embedding a few GIFs would have been a somewhat easy win and resulted in a higher hit rate.
For the next project, I should aim to have a solid preview build ready well in advance with the caveat that certain features may be added or polished before launch. I should also look to contact and get the press interested at this point too!
Things I’ve Learned:
- Continue to prototype gameplay early and rapidly, but don’t forget to do a full level to final visual quality for estimates!
- Build on the accessibility options from this game and better control the next game’s difficulty curve.
- Develop a UI framework ahead of time for the next game rather than make… whatever monstrosity I produced for Bubbles. Yuck.
- Better marketing messaging - be more clear with what kind of game I’m making, get copies into the hands of content creators sooner, get the Steam page up earlier.
What’s Next for Team Cats & Bears?
My next plan is to do a bit of freelance work to build up the coffers and supplement the income I’ve made from Bubbles and then prototype the next game. Once I’ve got something visually presentable, I’ll be happy to show you all!
Thanks for reading! I hope this has been useful. If you've been linked this outside of Steam and want to check the game out, here's a handy little widget to do so:
https://store.steampowered.com/app/945580/