Postmortem - Swing Racers

Introduction

My name is Cory Spooner - I am the founder of Morepork Games (based in New Zealand) and developer of Swing Racers. Prior to indie dev, I had worked commercially for around 10 years: at 2K Australia (Tech Art) on Bioshock Infinite, Borderlands The Pre-Sequel & The Bureau XCOM Declassified; at Sidhe (Designer) on Jackass: The Game, Speed Racer & Gripshift (360 version); Perception (Level Designer) on Stargate SG-1: The Alliance (unreleased). Before that a lot of hobby/community work with the Unreal Tournament series, since 2000.

Swing Racers is a 3D racing game on iOS where you pull your car around the house with a rubber band - the idea came from a dream I had, but also frustration with the controls of existing mobile racing games (not a fan of being given control sticks on a touch screen, or racing games where you just steer without control over speed - longed for something where you could intuitively do both easily).

Here’s a bit of a write-up on the things I’ll try to do differently on the next project, based on notes I took during development. They’re in no particular order:

1. Get help?
2. Analytics
3. Early previews
4. Feedback from early previewers
5. Attend a conference
6. Development blog
7. Allocate more time to marketing
8. Get the promotional material right
9. Plan for post-release updates
10. Make promo art ahead of time
11. Less text
12. Simple UI
13. More feedback
14. Be a programmer. Or just hire one.
15. Localization done properly
16. As many platforms as possible
17. Upload your own videos
18. Don't release during a major games event
19. Website domains and Facebook pages
20. Use a service to post to all social networks
21. Better-worded press emails
22. Proper email domain 

Get help?

Let’s face it - although there are advantages to doing things on your own, a whole game (of decent quality) is a ton of work. It made sense to do Swing Racers on my own (besides music) because it was a spare time project outside “real work”. The next project will be on my own too for the most part, but not because I think that’s the best way to develop. Money is the issue, of course.

If I did have some funding, and made Swing Racers again, the hiring priority from my point of view would be:

This is taking into account my own skill set, and would probably shuffle around depending on the type of game (Swing Racers had no animation, for example, which would be my first hire otherwise).

Analytics

Many times since release I have wished that I knew things about people playing my game - how many players chose the “left-handed” option? how many strived to unlock everything? which levels did they go back to the most? Knowing these things would help make better decisions on features for the next project.

One of the common analytics solutions would have been good, and the reason I didn’t end up having any was ultimately due to the engine choice (Flurry was integrated into UDK but was broken early on and never fixed). Having leaderboards in the game did give me *some* information, which gives me an indication of e.g. what percentage of players finished the last track:

Early previews

I held most PR back until release for Swing Racers (mainly because it was tough to get much at all), so that the little attention I could get was used to push up my rankings in the first week.

Ideally next time I’d like to do an early preview push that is *really* impressive, *really* clear that it’s a preview, and is well-endowed with links to follow the game (social networks, news subscription, etc). It helps a little bit that I have some facebook and twitter followers already this time.

Feedback from early previewers

With Swing Racers, I gave early builds to a couple of places so they could preview the game. Generally the feedback I got was good. One review site in particular said it was cool and that they didn’t have any gripes apart from the controls being a bit weird sometimes. Then when they game was released, they did a review and it were like “The game is OK but I get motion sick playing it. 5/10”...

Considering it’s the most prominent review available that was kinda annoying.

Next time I will push harder for any early preview feedback, so I can make sure I address the critics’ most grave concerns before they do the all-important review on release!

Attend a conference

I think it’d be wise to take my next game to a conference like PAX Aus, or at least attend one. I’ve been in the industry for 10 years and still never once been to a games event (not even in Aus/NZ)!

I do have a lot of contacts already and these events do cost money to attend, but it still feels like a no-brainer that I should be present there at some stage.

Development blog

Development blogs are a good way to keep people interested in what you’re doing, and another way to gain/keep attention. Besides which, they are a good outlet, particularly if you’re working on your own (the sheep in the paddock next to my office just don’t seem interested in what I have to say).

So for the next project I’ll do a development blog.

Allocate more time to marketing

Swing Racers was the first time marketing for me - I’d read a few guides about it and kinda had an idea of what to expect but I was still fairly naive. I’ve learnt that it is indeed a full-time job - there is always more marketing that can be done!

The *minimum* I will allocate next time, would be 3-4 weeks before release and 2-3 weeks after. That’s solely dedicated to marketing & promotion - including preparation of marketing material, trailers, screenshots, emails to press, researching and gathering contacts, building up interest, etc... The time *after* release is arguably more important, since your game is live and you need to keep your rankings up if you want any hope of being seen.

Get the promotional material right

You’ll see mentioned in various places that screenshots are super important. They are! As are the trailer and promotional art (app icon or box art).

What’s even more important is getting them right the *first time*. You end up sending them to press, posting them on forums, social networks, uploading them to game directories, etc. Getting them right the first time is even more important when doing localized versions of your game description (uploading 5 new screenshots at 5 different resolutions for 28 different languages (700 uploads) on iTunes Connect and arranging them - arrrgh).

I rushed the screenshots for Swing Racers a bit (plus I’m not great at taking good screenshots anyway), and my attempts at “App Store-esque” overlays to help sell the shots, were even more average. I lacked the motivation to change them all after the storm.

I also rushed the trailer. Next time I would like to plan it properly and spend more time on it (or even better, pay a pro if costs allow).

I feel like I did an OK job on the app icon considering I’m not that comfortable with that kind of stylized art, but I won’t forget to make that a priority next time, as well.

Plan for post-release updates

Something else I’ll do next time is a *solid plan for updates*. I have learnt that I should:

Make promo art ahead of time

If you get asked by Apple to provide promotional artwork, then you better damn well have it ready on time (you don’t get long). Here I was on my grandma-in-law’s dinner table, on my laptop, trying to concentrate on making an awesome, attention-grabbing, eye-pleasing banner to suit Apple’s tastes and specifications. Would have been tough even in the right environment, let alone under pressure.

I’ll be doing this ahead of time, next time, just in case.

Less text

I was trying to avoid having text in the game, but at some point in development I got lazy and apparently threw text all throughout the game. Then I decided it’d be a good idea to localize in 12 languages (which it is). My localization was generously done by friends this time around - I could’ve made it a lot easier on them by designing things a bit better.

Next time I will aim for as little text as possible, particularly stuff that’s involved in UI, like buttons etc, because they can be a pain to fit the different languages in e.g. “Race” in English, is “Autorennen” in German.

Simple UI

UI sucked a disgusting amount of development time. I think even when you have a good UI system it can be a pain, but I was manually positioning, scaling and animating *everything* in code without any live preview (the only other alternative in UDK was Scaleform, which I’d read performed really bad on mobile and would’ve meant learning Actionscript as well). Change position, compile, run, check, go back, change again, etc.

Having a cluttered UI also meant it was a bit hard to make it look pretty.

Next time around I won’t have to manually code the UI like that, but I will also put a heavier focus on intuitive design and try to keep UI as simple as possible.

More feedback

I had maybe 50 people try the game in it’s early stages which isn’t too bad considering it was a side project to begin with. In my experience that’s more user-testing than AAA games get in their early stages. Watching people play the game (without you telling them how) is the single best way to refine your game.

However, I struggled to get much testing done when I had a more finished product (beyond getting your friend’s kids to play the game) - I mean the kind of testing that finds bugs, not suggests new features. You tend to jam a lot of stuff in at the very end and that’s when you’re likely to introduce bugs, particularly if you’re a novice programmer. You also want to test on a variety of devices which I definitely did not do enough of. Despite the lack of later testing, Swing Racers ended up relatively bug free... I think it helps when you enjoy playing your own game.

So I will definitely try to save some testers for nearer the end. You can never have enough feedback.

Be a programmer. Or just hire one.

So many times in development I had the “oh sh*t” moment where I realized I’m not a real programmer and am in over my head. Amateur coding might suffice for something small but mine got big... like 10x bigger than I expected, and it becomes hard to manage if you didn’t do it right. There’s a reason them programmers are so anal about their standards. I got lazy and paid the price many times - I *just* scraped through and feel very lucky that I managed to get the game out there.

Ideally I’d like to have a programmer on board if I’m doing anything like that again, because I don’t think I have the chops or time to become a real programmer (a real programmer means experience). For now I’m aiming a bit smaller with my projects (particularly since I’m moving onto UE4 and don’t know C++).

Localization done properly

As mentioned earlier, I had very generous friends who did my localization for Swing Racers. I’d like to keep my friends, so next time I will try to design for minimal text that needs to be localized and/or have it done professionally.

As many platforms as possible

It’s become pretty clear that to gain any real attention, you realistically can’t just release on one platform (particularly one mobile platform, as with Swing Racers). The moment you are releasing something (of decent quality) on console or PC is when you get the attention of websites that people actually read, and people are more likely to take your game seriously.

I now feel like, the more platforms you are on, the more places on the web you can get reviews or mentions, and the more exposure you get (so a PC release may be worth the effort even if you don’t get the sales there).

So, although several platforms just may not be feasible without hiring help, I will definitely be aiming for it.

Upload your own videos

You don’t want people looking up e.g. “Swing Racers GAMEPLAY” on youtube and the top results are crappy videos that don’t show your game in a good light.

For the next project, I’ll first endeavour to have a game so good, that it doesn’t matter how it’s displayed (it looks good in any light). And if I can’t do that as well as I’d like, then I’ll endeavour to have my own videos near the top of the results so people see my “official” videos first, showing succinctly why they should be downloading my game right now.

Don't release during a major games event

Releasing during a big games event like E3/PAX/GDC means reviewers/journalists of any consequence are busy and will probably ignore your emails. And for the ones that don’t, the reviews or mentions from them are likely to be drowned out by the flood of big game news.

I released smack bang in the middle of E3. I did have a choice, but I had a feeling I was going to get a feature on release. I probably still would have got the feature later anyway but I didn’t want to risk it, because a feature is *everything* (plus I figured the amount of publicity I could get from press in the case of Swing Racers would be minimal).

Anyway next time I will plan the release date well away from gaming events.

Website domains and Facebook pages

It took me several weeks to realize I didn’t register www.swingracers.com . Seems like a small thing that might not help much but it’s a small price to register a domain, so why not?

I opted for a Morepork Games Facebook page but skipped having a Swing Racers one (didn’t feel like it’d be worth the effort and I’d just spam people more with the same stuff and piss them off). But I guess if you had a bigger game that’s likely to have more people searching for it, you’d want to make sure you had that page there.

Would be curious to hear what other people think about that one.

Use a service to post to all social networks

I attempted to custom write my news updates for each social network, but gee it gets out of control when you have Facebook, Twitter, Google+, LinkedIn, Tumblr, Flickr, instagram, youtube? vimeo? pinterest?

I’ll utilize one of those services that lets you post to all your social accounts automatically, in future.

Better-worded press emails

I was quite careful and paid a lot of attention to detail when writing emails to the press (did a lot of research and tried to personalize them a bit), but in retrospect, I think they might have been a bit too “salesperson’ish” e.g. “with controls that thetouch screen has been waiting for!”.

I got barely any responses from anybody, which could just be because it’s an iOS-only indie game, but I think I will try for a more casual, personalized, matter-of-fact wording next time.

Proper email domain

All of my emails to press came from a gmail address. I wonder if I might’ve got more responses had I sent them from my company website domain? It’s certainly appear more legitimate and professional, and it doesn’t cost much to get email added on to a domain deal, so next time I won’t leave myself wondering.