Let's say I have an app, and I need to request a permission from the user to use the main use case of the app. This permission request happens the first time that the user opens the app.

  • Scenario A: I display a prompt "in-app", saying "We'll need this permission to do this and that", giving context and when the user clicks "OK", I show them the iOS prompt for the device permission.
  • Scenario B: I display the iOS prompt directly on start-up.

I know that each mandatory prompt you display on first start will cause you to lose some users. My question is: are there scientific studies, or trustworthy results, who indicate how many are lost by each of the two options? Does scenario A lose more users because there are two prompts? If so, how many?

Note that I'm only interested in the user loss rate. This question is specifically not about good practices or user happiness.

1 Answer 1


I don't know of any study regarding how many users are lost permanently, but I've seen statistics for the number of users who declined the permissions.

The most detailed example is this article from the designer at Cluster.


  • The headline figures are that 30-40% of users accepted permissions when asked the first time they load up the app.

  • By educating the user before asking for the permission, this
    increased acceptance to 66%.

  • By only asking for the permission when the user triggers the
    which requires it, their acceptance rates were nearly
    at 100%. This was attributed to the user better understanding what
    the permission was for.

  • The last point is exactly how Android reworked the persmission system lately (since v. 6 I believe). Only asking for permission when the user expects it makes them think "well yeah, this is reasanoble" instead of "why do you need this?"
    – Big_Chair
    Commented Jun 7, 2017 at 12:15
  • The question I'm more interested in is: assuming the user is educated (for scenario A, in the in-app prompt, for scenario B, in the iOS prompt), would solution A (which allows for better "education", since you can add graphic elements, display account name, etc.) lose more users than solution B. But nonetheless, thanks! This is useful info.
    – Ted
    Commented Jun 7, 2017 at 13:07
  • 1
    Please correct me if I'm wrong, but your Scenario A is the same as the "Benefit Explanation" method in the Cluster article (66%), and your Scenario B is the same as the "Initial Blitzkrieg" method (30-40%).
    – user101673
    Commented Jun 7, 2017 at 15:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.