Setup your first Play Store Listing Experiment
Since Google I/O 2015, Play Store gives developers an opportunity to optimize the listing of theirs app by running A/B experiments. It becomes really important to have the right description of your product in order to increase its install rate. Several good applications have a visitors to install ratio of 30% and experimenting on the details of the listing helps your product achieve this conversion rate.
Here are some basic guidelines to set up your first Play Store experiment:
You can start your A/B testing from the Experiments menu from the developer console.
Type of Experiments
First of all, chose the type of experiment you want to run. Play Store allows you to run Global and Localized Experiments. You can run global experiments only on graphics — icon, featured graphic, screenshots and promo video. If your app is localized in various languages, you can run Localized experiment for short and long description, with one language at a time.
Name the experiment to differentiate all the experiments you run going forward. Choose your audience distributions for the variant(s) you experiment on. You can create upto 3 variations for each experiment. If you have less number of users, use one variation at a time with equal distribution of audience.
Choose the attribute
There are 6 attributes on which you can run Store Experiments on — App icon, Featured Graphic, Screenshots, Promo Video, Short Description and Long Description. Depending on which type of experiment you have chosen, you will be able to choose the available attributes. Choose one of these attributes and add relevant content for the same. For example, if you choose an app icon experiment, upload the new variation of the app icon.
Once you have added the content for the variation, set a name for the variation. This is relevant when you have added more than one variations for the same experiment. Once this is done, tap on Run Experiment and you have created your first Store Listing Experiment!
It is recommended that you run one experiment at a time to see the maximum result of any single variant. Also run the experiment atleast for 7 days or till you see enough data to view scaled installs.
Hope this guide helped you set up your first Store Listing Experiment. For any help, you can comment below or write to us on firstname.lastname@example.org.