Google Play Is Better Than The iOS Store (In Our Case): A Dangerous Post-Mortem

Purpose

In the 6 months since our game has been live at various app stores, we've gained valuable experience that other developers may find useful. Some of it is common knowledge while others may surprise you. In particular, there are some myths about Google Play (formerly known as Android Market) that we want to share. While our experience may not be typical, I hope it at least sheds a little more light on the mysterious workings of the different app stores.

History

Binary Helix is a tiny two-man company. We fortunately have a separate web publishing and software development business that ran on autopilot and paid most of our expenses during game development though I did go into debt to finance the shortfalls. We were able to devote most of our time to making our game.

Dangerous is the name of the game and the main character. The game itself is a single-player open and living universe 3D space combat RPG with an epic story. It's as if you mashed EVE Online, Elite, and Freelancer while adding elements of X3's empire building and Mass Effect's dialog and sidekicks. As you can imagine, it was a hugely ambitious first commercial game project for our studio—which in hindsight I'd tell aspiring game developers not to do. Even more, all this was to run equally well on mobile and desktop. I initially thought it was going to be a six month project, possibly a year in the worst case scenario. But since this was a project near and dear to my heart, feature creep inevitably crashed the party.

Dangerous took us about 2.5 years of development for initial release, with another six months of ongoing fixes and additions to reach the current version 1.1.0. So it's been a harrowing but exciting three year journey, and there are some lessons we've learned by releasing our game both on iOS, Android (Google Play and Amazon), Mac (Mac App Store and MacGameStore.com) and PC. We also sell the desktop versions directly from our website at whoisdangerous.com.

Versions

Initially, there was only Dangerous and the Dangerous Lite demo versions. These are the SD (standard definition) versions. When we released 1.1.0, we also added Dangerous HD and Dangerous HD Lite versions. On mobile, Dangerous is currently priced at $4.99 and Dangerous HD at $7.99. The HD version is currently only available on Google Play and on our website for desktop, but will soon be available on iOS, Amazon App Store, and Mac App Store.

Consumer Expectations and Sales

On Google Play we've kept the price of Dangerous at $4.99 since the launch. It's more of a niche game that has the depth of desktop games, and higher prices are generally justified due to the expected lower volume of sales as compared to a casual game. We made no real marketing effort prior to releasing the game on Android (my reasons are explained further in the document). We were lucky that a fan who bought the game notified the droidgamers.com website which posted a quick news blurb. Various other news aggregators picked up their posting and more or less repeated it, but the popularity of those other sites seemed mostly insignificant. From this, we received about 70 daily sales that petered out over a couple weeks down to the low teens. But sales grew steadily from there to around 25 sold per day.

So far, no players seem to be clamoring or waiting for the "inevitable" price drop as on iOS. Even on the free Lite versions, none have complained about the full version's price tag. As an experiment, I dropped the SD price down to $2.99 for a single day to see the effect on purchases. No change was observed. I was expecting a rush of buyers from app price monitoring consumers, but perhaps Dangerous just wasn't on anybody's radar yet. Or the price watching phenomenon does not seem to be as popular on Android as it is on iOS.

On the Amazon App Store, the daily sales were typically in the low single digits at best. Discovery is hard, and without some initial seed of interest, the game appears to be languishing there. Perhaps our upcoming marketing blitz will help.

Android lesson here: price drops don't do much for niche games like Dangerous. People are willing to pay for deep games. The biggest factor in our success was achieving consistent sales until Google's algorithms kicked in to link purchasers and viewers of Dangerous with other games.

On iOS, we launched with a "sale" price of $4.99 and raised it to the regular price of $6.99 a month after launch. We were loath to drop the price since we felt our game was rather niche and we did not want to play the race to the bottom game with price. However, once daily sales hit the single digits, we were left with a stark choice: stick to our guns and possibly "starve," or do some experiments with price to see how the market responded. The price drop back to $4.99 was well received, but in time, sales slowed.

The first time we dropped our price to $3.99 from $4.99, there was a large influx of buyers, but that quickly went away after several days. Due to the incredible number of apps appearing on the iOS App Store, players routinely rely on price watching websites. The lowest price we've set on iOS is $1.99 which showed the same short-term effect. We've since raised the price and left it at $4.99. At some point in the distant future, we may do $.99 or free on iOS, but only to promote some other related product and/or to build the fan base, perhaps when the sequel is out.

iOS lesson here: price drops on iOS will get tons more downloads from the price-conscious consumer, but the effect does not last long. I surmise it's a better long-term strategy to get noticed via reviews or product updates which generate a more lasting sales effect. We also never seemed to achieve the critical number of purchases needed to be reflected in the "Customers Also Bought" list with other well known apps. (Perhaps making the game free or a buck for a week would do the trick.)

We released on the Mac App Store with high expectations. After all, a game of this complexity is better suited to desktop systems. It was a resounding dud. About a day prior to the release, we submitted a press release via prmac.com. We did receive one review on oneclickmac.com with a poor 2 out of 5 rating, but I felt the reviewer didn't really give the game a fair shake, and we had a lengthy email exchange to their credit. Sales were in the low single digits. Even though we were on the first page of the RPG and Simulation subcategories (and reached nearly top-ranking positions), sales were virtually non-existent. Daily, there were hundreds of downloads of the free Lite version, but very few translated to actual sales.

We were not featured by Apple, so have no clue on how that might have affected sales. Price does play some consideration. We felt our game was comparable to other games at the $19.99 level, but got few sales there. Only when we were sub-$10 and $5 did sales noticeably pick up to a handful a day. No hype from an unknown studio apparently equals no sales.

Months later, we featured a substantially discounted price when we launched on the MacGameStore.com, but so far, sales have not really impressed. Much like the Mac App Store, I believe you need a certain level of hype and marketing to succeed. As we now ramp up the marketing with version 1.1.0, I hope to see improved sales through those two channels.

Mac App Store lesson: getting a high rank in a subcategory will not automatically result in more sales. Even price drops resulted in less than stellar sales. Even hundreds or thousands of downloads of a free Lite version did not result in more sales. This must be coupled with additional marketing at the very least. Or if you're lucky getting featured by Apple.

You're doing it backwards!

Yes, I realize we should have started the marketing engine long before initial release. I'm hoping the strength of the product will pull us through this next phase. Also, I felt that the new 1.1.0 update finally brings enough UI simplification and graphical beauty that it was worth holding off so as to present Dangerous with the best possible first impression now that we've been able to incorporate feedback from the early adopters. We'll see how the folly of my thinking plays out.

Android Is Too Fragmented

There is truth to this since the number of Android devices, not to mention the myriad of installed OS versions, is absolutely bewildering, and a small developer has zero chance of testing on most of them. So because of Android's perceived and real fragmentation, there is a reluctance from developers to enter the fray. To some extent, this has helped us since there are fewer competitors who have spent years developing their game. So the attention to detail and relatively stable releases of Dangerous help it stand out from the pack.

The fragmentation issue can be mitigated by using a cross platform framework like Unity. Not only is porting a snap, you can leverage Unity's developer support community to solve your fragmentation woes. You are more dependent on Unity, but a source license is possible if you have the funds and inclination. We have neither.

Also, it is critical to provide free demo versions for people to test on their devices if you're planning on selling at a premium price. We have both free SD and HD Lite versions for people to try before buying. This catches most of the support issues because people are not usually willing to shell out $7.99 for a mobile game without making sure it runs on their device. This required level of commitment also reflects favorably in the review score for Dangerous HD which sits at 4.5 stars.

Google Play Has An Annoying File Size Limitation

The biggest technical issue we ran into was Google Play did not let you upload binaries larger than 50 MB, and supposedly, some phones can only handle a 30 MB download. Dangerous version 1.1.0 is about 200 MB and Dangerous HD is 500-900 MB depending on platform and/or device. This meant breaking up the binary into an executable and a data portion which violates the raison d'etre of using a portable framework like Unity. Luckily, we found a user contributed solution that automated this, so we didn't have to re-architect our game just for Google Play.

(Note that the Amazon Android App Store doesn't have this file size limitation.)

Unfortunately, that solution initially required hosting the data on our servers. This was exacerbated when we released the HD version with hundreds of MB of additional texture data. The bandwidth costs of hosting the data on Amazon's cloud ate into our profits by $600-$800/month. This was doubly bad since pirates and the free Lite versions were now actually costing us money. Only recently were we able to migrate the data to Google's servers.

Support costs on Android

On the iOS App Store, Apple handles all aspects of the sale including any returns. On Android, you get a transaction receipt for every purchase including the email and name of the buyer. You're also expected to deal with any dissatisfied customers directly. This additional support does add up, and is an unfortunate opportunity cost for the small development team.

This was the primary reason Mika Mobile, the developers of Zombieville, USA, cited for stopping further development on Android. They were making much more money on iOS without the support hassles. For less well known developers like us, this is simply a cost of doing business on Google Play. Also, in a sense, it gives you the opportunity to interact far more closely with your customers and possibly create life-long fans.

We have a permissive policy of honoring any refund requests even after Google's 15 minutes window. Players can simply email us. We try to resolve their issue, and if all else fails, we happily refund. We have not found players to abuse our good will.

Myth: Android Piracy is Rampant

Actually, piracy may be rampant and not a myth, but from our experience, it is no worse than on iOS. So for all intents and purposes, at least for us, it is a myth or at least something that shouldn't concern the average developer.

When any of our games start up, they grab a message of the day (MOTD) from our server. This gives us a way to talk to our players to let them know about new updates and to ask them to review our game, etc. We're also able to gauge a rough estimate of how many players are playing daily. Early on, there were about 3,000 unique players a day, and it has since dropped to about 1,000 a day. (Note that we aggregated by IP, so those numbers are probably low ball due to IP sharing/NAT by ISPs.) There were two huge spikes in the tens of thousands by networks located in Russia and China which didn't see a corresponding rise in sales, so those were when piracy was at its peak. However, the use of these stats are unreliable for determining piracy, and it's possible the real numbers are much higher. I don't really care, honestly. The numbers that matter are the revenues, and those have been trending up.

Here is an actual email we received:

"Subject: Dangerously Amazing

I'm ashamed to say I downloaded a cracked version of Dangerous, but was so impressed with the effort, depth and quality of the game I had to immediately buy it.  Most other games lose their novelty after 2 minutes which is really not worth more than a free flash game to me.  I look forward to buy any of your future products. "

So there's no denying that piracy exists. But the fact that customers are willing to pay a premium price after trying the demo or a cracked copy belies the claim that indie developers can't sell a game at a premium price on Android. Would we earn more if we could reduce piracy? Probably, but piracy also exposes your game to many more potential customers. So why fight that arms race? We prefer to focus on improving the game for the paying customers.

Also, there are different ways of tackling piracy without encumbering your customers with restrictive DRM (digital rights management) in software. Another well-known desktop indie game called Space Pirates and Zombies displays a note to the player upon start up that essentially implores players not to pirate. This was a little too in-your-face for me since I feel that legitimate players shouldn't have to see that (it's akin to the FBI warning on movies albeit on a subtler level in my opinion), and I opted instead to beef up the credits section with a large photo and text to inform the players more about Binary Helix and our external collaborators, who are composer Sean Beeson and artist Kip Ayers. To let players know we're just a couple of quirky guys trying to make an honest buck after investing years on the project. To humanize the developers and contributors. No idea if this worked in any real sense, but I felt more comfortable with this approach. To me, every paying customer is a prize, and I do not want to subject them to any hassles.

In fact, our Lite versions are rather permissive in that you, in theory, could complete the game. The whole game is included in the Lite version. They only restrict the amount of XP you can spend. This restriction potentially makes for a very difficult game if they venture to the lower security areas, but a clever player could use many game mechanics to still succeed akin to an "Ironman challenge" play through. Lite versions also have the cloud save/load feature which might be abused in a similar fashion. But as far as I'm concerned, players can play however they wish even if they get their kicks out of beating the developer imposed restrictions.

In addition, we have been constantly improving and updating the game since release. This makes the official versions a better value proposition than players having to hunt down pirated updates all the time.

Myth: You Have To Price Your Game For Free Or $.99 On Android And Use IAP To Thwart Pirates

Dangerous on Google Play costs $4.99 and never really had a price reduction since release. Dangerous HD costs $7.99 and now accounts for 2/3 of the sales. We're a completely unknown studio, and our HD game costs more than most premium games by large publishers and developers such as Gameloft, EA, Madfinger and even more than Minecraft: Pocket Edition. This was not intentional since I merely set the HD version at what I believed was a "fair" price. Of course, our sales are nowhere near those other games' sales numbers, but we're set to make $4,000 after Google takes its cut for July. Considering we have no PR, no marketing and no reviews, I can't really complain.

Is piracy really rampant at 90% as others report? I don't know. But I don't think so. Most pirates would not buy our game if it were priced at a dollar, so using such spurious statistics feels deceptive. Our focus is on our customers, and we are adding language support in an upcoming version. This should hopefully increase sales in big foreign markets like Germany, Russia, China, etc. Our way to fight piracy is to constantly give the players what they ask. And update early, update often. I think customers will gladly support such companies rather than rely on dubious pirate sites which may include malware.

We are going to add IAP (In-App Purchase), but only for vanity items and credits, and for those who want to support Binary Helix. The game can be completed already without the need for any IAP. You'll be able to buy a ship with a different color scheme but is otherwise identical in function to ships you can obtain in the game. In addition, we're going to add advertising to our free versions as a test. Currently, the free Lite versions have no advertising with only a simple restriction on how much XP you can spend.

Myth: You Need Ongoing And Constant PR To Get Sales

This is one of the more surprising lessons we've learned from Android. Other than a single news mention in droidgamers.com, which resulted in a peak of 70 purchases in a day, we have had zero reviews by the "mainstream" Android review sites. This is partly by design, and partly due to Google Play not making it possible to give away promo codes like on iOS. I wanted to hold off on doing a major PR push until the game was in a state I was proud of (it's the perfectionist in me, I suppose).

(Why release in the first place? Simply, we ran out of time and money, and felt that the game was good enough after 2.5 years of development. We constantly deluded ourselves into thinking it was always a month from being completed.)

We received a lot of feedback from players, and I felt that the game could have and should have been even better. It's also a labor of love. So we took all the feedback and criticism to heart, and about six month later, we released version 1.1.0. This is the version I'm finally proud of, and will try to submit it to the more popular sites for review. I'll do a follow up post when those numbers are in.

So how did we get a steady stream of customers with no special mentions by Google Play, no advertising, PR, or social media campaign? Mostly word of mouth, and as far as I can determine, the automatic links provided by Google when people view or install other similar games. It took a few months of somewhat sporadic sales, but there came a time when enough people viewed or downloaded the game that we were listed in the "Users who viewed this also viewed" and "Users who installed this also installed" areas.

Some of the more popular games which algorithmically linked back to us are: Dead Space, Dead Trigger, Earth and Legend, and Mass Effect: Infiltrator. These other games have far more installs and a larger customer base. There is no easy way (as far as I know) to figure out which games link back. Oddly, the games that are linked from Dangerous don't seem to link back to our game. 

But don't let this myth be your guide. Lesson here: While it's always better to build interest with social media and get previews and reviews from as many sites as possible to increase your chances with financial success, not doing so is not an automatic death sentence. Good games will always find an audience sooner or later. And currently, there is less competition on Android as on iOS, so quality games naturally stand out more.

Myth: Support Costs Render Android Untenable

We're a two-man studio, and I handle support so as to let our coder do his development wizardry. When the game launched, we received a number of support emails, probably 2-3 a day. This has dwindled down to one every 3-4 days. In general, if I'm not able to resolve the problem they're experiencing, I'll happily offer a refund. Our app description page at Google Play tells them to contact us for a refund after Google's 15 minutes. We have not found players to abuse this, and have only had a handful of requests. Those that received the refunds were thankful (some even surprised we kept our word). Lesson here: trust your customers and treat them well. You'll also have fewer support emails if you use a best of breed platform like Unity which lets you quickly test without needing a compile cycle.

Areas Where Google Play Clearly Beats The iOS App Store

Caveats

We have not released the latest update version 1.1.0 including a new HD version to the iOS store due in part to the laggy iOS review process, so it's possible that it could be better received than Google Play. Especially since we are gearing up to finally push for reviews, etc. However, even if that were the case, it still does not diminish the fact that prior to the HD release, our SD version was still gaining traction and consistently selling more copies on Google Play than on iOS by a factor of 3x or better.

Another factor that may play a large role is that the closest game competitor (which is Galaxy on Fire 2 by Fishlabs) seems to be available to fewer Android devices than Dangerous. For example, GoF2 is inexplicably unavailable on my HTC EVO 3D and Kindle Fire, and it did not successfully install the data portion on my Android-enabled HP Touchpad. (Dangerous works on all three devices and is also available in the Amazon App Store.) On iOS, the situation is reversed, where Dangerous only plays on the later generation devices due to needing greater background processing with the living universe.

Also, while I take great personal satisfaction in earning upwards of $5K each month total from a game I've always wanted to make, we still have not come close to breaking even on the three year development costs. And the revenue still doesn't pay the full monthly salaries and business expenses. But many roads lie before us in monetizing opportunities (such as overseas app stores and digital distribution on PC), and we have hopefully established ourselves as a competent micro studio specializing in bleeding edge desktop-quality games for mobile. I will also be reaching out to traditional publishers for other platforms.

Of course, you'll want to take all this with a grain of salt as your mileage may vary. We may have just been lucky, and the numbers may go south tomorrow. The anticipated PR blitz may not move the numbers much farther. However, I believe players will recognize a quality product that comes from the hands of people passionate about their game. The one recommendation I'd make to any new studio is to cut your feature list in half and polish the heck out of the game instead.

Conclusion

So those who dismiss the Google Play store out of hand due to rampant piracy or to the notion that an unknown developer cannot compete with established brands are clearly mistaken as sales of Dangerous has demonstrated. Android players are clearly willing to pay for a quality experience as evidenced by our $7.99 HD pricing which costs more than virtually all premium games from major developers/publishers. While piracy exists on all platforms, we don't waste any time trying to fight it. Instead, we are focused on providing paying customers the best experience possible.

In the end, though it took three years, I'm glad we have such a deep and immersive game for mobile and desktop under our belt. (But the next projects for Binary Helix will definitely be shorter in scope—fingers crossed. Though we have some great ideas for future expansions as well... Uh oh.) Our naive approach to marketing has also given us surprising and instructive results that shatter some well-worn Android myths. It'll be interesting to see how the landscape looks after a hopeful PR and marketing blitz campaign.