Store listing A/B testing
In Google Play Console store listing experiments, test one element at a time (icon, feature graphic, screenshots or short description) and run it for at least 7 days, ideally a full week. If you change two elements at once you cannot tell which one produced the lift.
One variable at a time
Change a single element per experiment: the icon, the first two screenshots, or the short description. Change two and you will never know which one won, so nothing compounds.
Duration and traffic
Run for at least seven days so weekdays and weekends are both covered; with low daily page views, stretch it to two weeks. Experiments stopped early produce fake winners.
If the conversion gap is under one percent, do not ship it — that range is measurement noise.
What is worth testing
- Icon: the single highest-impact element
- First two screenshots: the frames shown in search results
- Short description: the first half of the 80 characters
- Feature graphic: only for featured placements
Frequently asked questions
Where do you run store listing A/B tests?
Google Play Console → Store presence → Store listing experiments. No extra tool or code required.
How long should an experiment run?
At least 7 days, ideally a full week — weekday and weekend behaviour differ.
How many elements should you test at once?
One. If you change the icon and the screenshots together you cannot tell which produced the lift.
When should you apply the result?
Wait until Play reports a confidence interval; experiments stopped early tend to turn noise into a permanent decision.
Compare your own app against this data
Enter a package name and we will benchmark its store listing against this sample. Free, no signup.