When it comes to maintaining two separate apps on two separate app stores, there’s no doubt that Google Play provides a lot more data to us as developers comparing to iOS.
Google Play Store Listing Experiments:
Two years ago, at some point in 2015 – Google Play introduced the Store Listing Experiments. And among the things you can test you may find:
Store Icon
Graphics
Screenshots
Video
Short Description
Full Description
Last year we tried a couple of screenshots and descriptions tests, but none of them made a huge difference. It became clear that we need to improve the CTR significantly from the store listings to get more downloads.
When the user searches for something on Google Play, he sees 4 main details before he chooses which app to install:
Logo (App icon)
App Store Title
The company behind it
The app’s rate
For a long time we just knew that our App icon could perform better.
Google Play’s Experiments allow you to choose a Global or a Localized test. Once you’re in, just choose how many versions you want for every test. You can easily stop a test and continue to the next one at any point. And I have to admit that it adds a lot of fun for both the Design and the Product teams to start a competition, maybe even bet on which icons will perform better.
If you’ll look at our old logo, you will notice that it’s missing a chart. An element that all of our competitors always had in their icons. And while we ranked very well for many terms (including#2 for “Stocks”), something in the icon simply didn’t work.
In the last 4 months, we had over 20 tests, between 3 icons each time, in 5 of our top countries.
Truth is that you can tailor the icon per GEO, but that means that the icon that will be installed to the user’s device will be different, as you can only tailor it on the store.
In the end we were able to find a logo that performs better in 4 out of our top 5 GEOs with an average of 12% improvement in CTR.
In an app with 8,000 daily organic downloads, that’s another 960 downloads every day, or 350,000 downloads a year.