Published in


The new Google Play policy: why you should stop using push ads now

At the beginning of this month Google revised its content guidelines for apps published on Google Play. At AppBrain we’ve always been great proponents of a good user experience, and in 2010 we had already launched a spam filter for the Android market. We therefore welcome the changes that Google has made to the Google Play policy, especially with regards to decreasing the annoyance factor of ad networks.

We wanted to know if Google has already started enforcing the new policy. The policy applies to all apps effective August 30th, but took effect immediately for new apps or apps that have updated since August 1st. We used data from our AppBrain app discovery platform, which is continuously synced with the Google Play market, to look at apps that were unpublished from Google Play over the last 2½ months. We found 59,737 apps that had been unpublished in this timeframe, on average 728 per day.

Note that the removals we count are not necessarily all by Google, as app developers themselves can also unpublish an app. In addition, AppBrain only keeps track of apps that are published in the US, so if a developer decides to restrict his app to not be available in the US we will count this as an unpublished app too.

The graph of unpublished apps over time is shown above, and it looks like Google hasn’t started strictly enforcing the new policy, or at least it didn’t lead to a noticeable increase in number of apps unpublished.

To get insight into which factors contribute to apps being unpublished, we coupled the set of unpublished apps with the presence of ad networks in those apps from our AppBrain library stats data. We also added basic market data such as the app category, number of ratings, average rating, and the date the app was first published. We compared the unpublished apps with all currently active apps to find factors that correlated with the unpublishing of apps. These factors were determined using a multivariable logistic regression model.

Naturally, such an analysis can only prove correlation and not causation. We corrected for the bias in the data that developers of inferior apps (copyright infringement, adult content, etc.) have a preference for particular ad networks. By including the ratings, number of ratings and the presence of adult words in the description, ad networks that mostly have badly rated apps don’t get assigned an unfair value in this model.

The results are summarized in the image below. Even though Google hasn’t changed their activity since August 1st, it’s interesting to see that the factors that were predictive of apps being unpublished were already mostly related to practices that are discouraged in the new guidelines: ad networks that do push notification ads, put ad icons on the desktop and use your contact data to send ads to your friends. Inclusion of some of these ad networks had an equally strong effect as the presence of porn words in the title and description, which more than doubled the likelihood of being unpublished.

The factors that were related to being active on the market were mostly indicators of serious apps, such as being published in the Education, Travel & Local or Lifestyle categories. (Gaming categories and Entertainment and Personalization were the categories with the highest unpublishing rates).

Having high market ratings and many market ratings was also associated with being active on the market. On average the odds of an app being unpublished increased by 7% per added ad network. The few exceptions to this negative influence were mostly trusted names like Google’s own Admob, Millennial Media, InMobi, and also our own AppBrain AppLift network. The influences of the most commonly present ad networks are shown in the graph on the side.

At the moment push ads are present in 14% of all newly launched Android apps on Google Play and ad icons in 10%. We hope that Google will enforce the new policy more strictly starting August 30th to make Android a better platform for users and non-intrusive ad networks alike.

What does this mean for developers of Android apps? In our view, now is a better time than ever to stop using ad networks that rely on push ads, icon ads or other forms of intrusive advertising. We now know that these forms of advertising have been associated with a higher chance of being removed from the market over the last 2½ months, and it’s likely that a big wave of market removals will follow after August 30th. In other words, push ads and other intrusive forms of advertising don’t provide a sustainable business model, as they chase away users and jeopardize the listing of your app on Google Play.
There are a number of great alternatives which can even be combined easily with mediation as described in our last post. The AppBrain AppLift SDK is built from the ground up to be user friendly, and provides a highly monetizing interstitial or a standard banner format that can be plugged into Admob and AdWhirl mediation if needed.

We’ll be keeping a close eye on the unpublishing rates and will report back here if anything changes after August 30th.

Happy app developing,

Mathijs for the AppBrain team.



The go-to place to promote and earn money for Android developers.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Mathijs Vogelzang

Building the top Android CPI network, AppTornado, AppBrain, Swiss Codemonkey