The one thing that's not possible to test is when a download is uninstalled, which affects this metric significantly.
I find that I install a lot of android apps, but uninstall the vast majority of them. My iOS device, on the other hand, gets fewer downloads, but virtually all are paid and virtually all have stayed for good.
After owning 3 android devices, iPod touch and an iPad 2, I've come to the conclusion that the iOS apps are substantially better. To balance the field I have to download many android apps to find the best ones, which has been a challenge because most of the android and amazon markets are complete crap.
One example... I use audiogalaxy for streaming music. The iOS version is perfection, it streams well, UI has no glitches that I've found and it uses the iOS dev patterns well. The android version has many UI issues, it's sluggish and buffering works, but not as well as iOS. In addition, the menu system is convoluted on android making it very difficult to navigate. This is just one example, there are many android apps that have these issues (same thing with beyondpod).
I am impressed with android numbers (those of us who bought G1's never thought this day would come), but I don't think it's translating into more sales. Android still has a long way to go before it catches up with iOS, imo.
Apple's marketshare for computers priced over $1000 is >90%, meanwhile the largest personal computer manufacturer in the world is looking to exit the business. I think this is a harbinger for what iOS and Android will become.
There is definitely some maturing required to occur on the Android side and I think this is represented by the continuing hesitation of pay app developers on the platform. The article didn't shed light on the revenue figures but rather concentrated on downloads only, which as per your experience can be misleading. Additionally the split between paid/unpaid and the average revenue on paid apps would be the meat necessary to attract developers. I feel this may be a long-running issue for Android as one of its primary attractions for users is the low acquisition cost, thus it's marketed to a consumer who is less likely to spend. (Seconded by the high piracy rates on Android.) This creates a cyclical issue, the android app store quality becomes diluted by lack of signature apps, so consumers become less likely to purchase apps from a store where they are accustomed to receiving poorer quality applications. So developers don't consider Android the first-platform of choice, thus perpetuating the cycle. For consumers buying become a risk exercise, thus pushing more favour to the ease of piracy, and developers to ultimately consider the platform secondary.
Actual Android penetration is also difficult to gauge, while iOS has predictable hardware, retail and pricing. Android-based phones have received significant sales promotion offers (e.g. 2 for 1 is quite common), the models available fit across a price and technical spectrum from top notch to crappy-phone-that-can't-do-much. I believe this leads to why app downloads are lower per device. (When the randomness of consumers should yield a similar app-download figure, perhaps even slightly skewed towards Android due to the larger amount of test and free software available.) Instead we see Android lagging despite having significantly more devices in the wild. (This doesn't build confidence for quality developers to make Android a priority destination.)
On the other hand Google's priority is to ensure ubiquity (whether good or bad phones), such that their ad revenue based services don't miss the smartphone revolution. Their strategy for this is effective, it's reminiscent of windows vs mac os. In the end iOS isn't Android's problem - Apple's iOS strategy will always be selling less units than Android, but as we've seen in computer manufacturers it may not be about how many you sell, but rather how profitable it is to sell at all.
Side note:
Unfortunately for everyone, this means there will always be Android/iOS flame wars in the same way we've come to expect Windows/Mac OS/Linux flame threads. The plus side is that innovation will be continuous for both platforms.
As an Android developer, I love this trend. It's amazing to be able to reach such a large and growing audience in the mobile world without having to live inside Apple's ecosystem. It's also exciting that the high-end Android phones out now are increasingly awesome. I hope this translates into an increase in paid apps' success, given that more users with more money seem to be using Android.
I'm not selling yet, no. I've released one app in the market (pressureNET, a crowdsourced barometer network for tablets) that I'm looking to monetize over the next few months and I'm hoping selling weather prediction apps will be possible :)
I'm still in the very early stages of the free app, though, so there's lots of room and little data from my end on selling paid apps.
No, that's not actually the case. Or at least, not from that report. The Millennial numbers are about one thing alone: the Millennial network. If we knew that the install base for Millennial-provided ads exactly mirrored the install base for smartphones, then yes, but: that's not actually the case, as different ad nets are used to different extents on different devices. For example, you could put out a press release trumpeting iOS 100% domination of mobile ads if you just used numbers from iAds...
It's not just a one time state that shows this. Millennial has been posting these reports every quarter or so, and I do remember at least a few of them when iOS was ahead. It's probably not completely accurate, but it does show a trend, from iOS dominating their stats a year or so ago, and now Android dominating them.
Considering many of the top grossing apps on Android (and iOS as well) are free with in app purchasing, I am starting to believe paid app downloads are going to increasingly become a niche market for apps.
I also wonder how ad-supported apps affect these numbers. If apps are free but ad-supported, there's more incentive to download and try them out, even if you only ever use them once and then delete them.
If true, this has been a while coming, but I'm taking the source with a grain of salt. It makes no mention of methodology for data gathering. Apple don't, in general, share their data. Nor do Google. It doesn't mention of what a "Mobile App download" is. Are iPod Touch and iPad included in the stats (which make more > 50% of iOS installations)? In app purchase data? What about Amazon App Store?
Of course, the only way to get the even skimpy details is to cough up a price (that isn't even available until you log in). The rest of their site seems to indicate that "Mobile" means phone handsets, which are not the only source of installs for iOS and Android.
Edit: I had a small rant about "analysts", the plague of our industry, but decided it didn't add anything (I hope I'm preaching to the converted).
It sounds like they are including third party app stores, presumably including Amazon's:
Being a free platform has expanded the Android device install base, which in turn has driven growth in the number of third party multi-platform and mobile operator app stores
I would like to see data on developer profits from paid/ad based apps though. I wonder if this may be due to more free apps on the Android Marketplace.
There are an infinite number of metrics to compare devices. But this is a pretty general one.
You could also do profits, users, developers, developer income, apps available. Give it up, it's arbitrary and you're not doing much more than justifying the conclusions you've already made about the platform.
The audience for this metric isn't consumers choosing whether to get an iPhone or an Android phone. It's developers choosing whether to support iOS, Android or both. For that audience, this is a very relevant metric. It's also a fairly objective one.
(Of course there's a second-order effect: users will choose a platform partly according to how well it's supported by developers. But anyone choosing which phone to buy now on the basis of this metric is betting on a longer term than they'll probably keep that phone for.)
I find that I install a lot of android apps, but uninstall the vast majority of them. My iOS device, on the other hand, gets fewer downloads, but virtually all are paid and virtually all have stayed for good.
After owning 3 android devices, iPod touch and an iPad 2, I've come to the conclusion that the iOS apps are substantially better. To balance the field I have to download many android apps to find the best ones, which has been a challenge because most of the android and amazon markets are complete crap.
One example... I use audiogalaxy for streaming music. The iOS version is perfection, it streams well, UI has no glitches that I've found and it uses the iOS dev patterns well. The android version has many UI issues, it's sluggish and buffering works, but not as well as iOS. In addition, the menu system is convoluted on android making it very difficult to navigate. This is just one example, there are many android apps that have these issues (same thing with beyondpod).
I am impressed with android numbers (those of us who bought G1's never thought this day would come), but I don't think it's translating into more sales. Android still has a long way to go before it catches up with iOS, imo.