18 Aug 2016

Interesting App Store Metrics

Some metrics are commonly quoted — that Google Play has a billion active users, or that Apple's iOS app store earns more money for developers, or that Windows Mobile is failing.

But do you know that the average Windows Mobile developer earns more than the average iOS or Android developer? The Windows developer earns $11K a month, compared to 8K for iOS and 5K for Android. Though Windows mobile has far fewer sales, it seems to have even fewer developers, so each developer earns more. Many apps seem to be paid on Windows Mobile, versus free on Android and iOS.

iOS apps bring in 75% more revenue than Android apps. But that's what users pay. If you look at ad revenue, iOS is only 17% more than Android.

How do upfront payment perform relative to in-app purchase and advertising? Here's the answer for Windows:

graph

Which categories have more downloads? Here's the answer, again for Windows:

1_downloadsbycategory

A more interesting statistic is which category has the most downloads per app. That's where you might want to concentrate your efforts:

3_totalopportunity.png

Moving on from Windows, I wish Apple released information showing the revenue per app across all their four app stores — iOS, macOS, watchOS and tvOS. And number of downloads per app. That way, developers can choose the platform that produces the most revenue, amongst Apple platforms. And the competition will hopefully drive down prices for users where they are high. If Apple hesitates to release the raw numbers, they can normalise it.

To take an Android example, a question like "How many users are there?" has a lot of nuance. To begin with, let's take Google Play's number of 1 billion active users. That immediately eliminates people who don't have Play (like Chinese phones or Kindles). Or who don't have Internet access at all, like the security guard at my apartment. Further, let's say you're planning on using an API that was introduced in Android L. That's 40% of Android users, which is 400 million. You might as well ignore the other users.

As another example, if you're making an India-specific app, the iPhone's market share is insignificant, like 2%, but if you look at phones priced at more than ₹20K, Samsung's share is 44%, and Apple's share is 28%. Go above ₹30K, and Samsung and Apple are roughly at par at around 45%.

There's so much nuance around metrics. Merely tossing out one number is not good enough. It will lead you to the wrong decisions. Make sure you're measuring the right thing.

No comments:

Post a Comment