Earlier today Flurry posted findings of a recent study of theirs in a blog post of theirs saying that revenue from the Amazon App Store is over three times what it is for an app in the Google Play Store. This is horribly misleading the public, and journalists alike to spell doom and gloom for Google. In fact, when reading between the lines on the study, there are many holes in their data collection that could result in such a headline selling study finding.
1.) Only in app purchase freemium type apps were studied
So saying "an app makes more revenue on Amazon" is a massively over generalized statement. We can modify the statement by saying "Freemium apps make more money on Amazon" because "Apps" in general is incorrect. The blog post "For Generating App Revenue, Amazon Shows Google How to Play" is wrong because there are multiple ways to generate app revenue and this is a single method. Not only that but they used a play on words with "Google How to Play" despite not measuring any data after the Google rebranding switch, that may not affect the results, but points out that Flurry was in it for a good headline.
2.) Sample size was never established nor data collection methods
In order to have faith in any study you have to understand you sample size and data collection methods to avoid bias in the data. Flurry simply stated that they watched "A basket of top apps" not stating which ones or what their function was. This is important because if they were all games for instance, then there may be some bias. Additionally if a "basket" to flurry is 4, then we have an error in sample size where an anomaly can easily invalidate the data. By not making this clear the data should not be trusted in any way.
3.) The numbers are per user - and the demographics are MUCH different
People that have Android phones often have recieved them unsubsidized. Additionally you can download free apps (exactly what these are) without adding a credit card on the Google Play Store. On the Amazon market you must enter a credit card. Additionally Amazon has a filtered demographic. They consist of two types of users; Kindle Fire owners, and sideloaders. The Fire owners are willing to purchase a device without a subsidy and are probably already well integrated into the Amazon eco system, this makes them more likely to buy. Sideloaders tend to be more technically savvy and for this reason they are more likely to adopt the idea of purchasing goods digitally. Due to this filtered demographic on a per user basis Amazon very well could make more money. That doesn't help however when you're comparing 10 people on Amazon to 10,000 on the Google Play Store. This study fails to point out that Google probably made WAY more money overall.
This is very clearly statistics being contorted in order to drive attention to Flurry. Shame.... On.... Them....