we would like to see which one leads to more downloading
First things first - any experiments you run are synthetic models. You can certainly try to predict how some people will behave under simulated conditions with A/B experiments like this, but you have to know that even if your experiment is internally valid and sound a "winner" emerges, it doesn't mean it's externally valid and can be extrapolated to predict the behavior of App Store users in the wild.
Unless you're actually testing a few variants live in the App Store, you have to take the results with a grain of salt because they're just a model, and models are really only as strong and reliable as the amount of time and money you're willing to cough up to make it more of a realistic simulation...in which case you just as well do the real thing.
You mentioned cost is an issue, so it's important that you keep in mind the limits of the method because confidence costs $.
That aside, your question of cause-and-effect for a single independent variable lends itself perfectly to the format of A/B testing.
Your particular use case is closer to marketing conversion testing, a niche subset of testing which has its own hordes of expert tools and communities that include Optimizely, Unbounce, even Hubspot.