How to calculate the engagement rate properly?

Hi there

I took a look at the mobile benchmark study that was published by Urbanairship a while ago and wonder how exactly the "engagement rate" was calculated. My goal is to calculate a compareable engagement rate  for our own apps on a daily as well as monthly level. Is this possible? If so, whats the correct combination of metrics to do so.

Furthermore I have a question regarding the Push Response Report. Are the Direct Opens and Incluend Opens being deduplicated when reporting a time period e.g. a day or a series of days when multiple messages are sent? I'm asking because I wonder to what degreee I can compare the Direct/Influenced Opens from the Push Response report with the Direct/Influenced Opens from the Message Activity Report of indivudal messages.

Thanks a lot
Claudio

Didn't find what you were looking for?

New post

Comments

4 comments

  • Hi Claudio,

    Yes, that would give you a reasonable percentage of opens that occurred as a result of notification activity out of the total opens that month.

    Thank you,
    Sean Conlin
    Urban Airship Technical Support
    Portland, Oregon

    2
  • Hi Claudio,

    This is Sean with Urban Airship Technical Support.

    The Engagement Rate in that study is based on Influenced Opens. For more information about Influenced Opens, please see our Influenced Opens Primer.

    Both Direct Opens and Influenced Opens can only be associated with a single push. So if you send multiple pushes to the same device during the 12 hour window that Influenced Opens are calculated, opens from that device will only count as an influenced open toward the most recently sent push notification.

    Please let me know if you have any other questions or concerns.

    Thank you,
    Sean Conlin
    Urban Airship Technical Support
    Portland, Oregon

    0
  • Hi Sean

    Cool, thanks for your feedback. So I could calculate compareable engagement rate as follows: Monthly Influenced Opens divded monthly (non-unique) App Opens or am I making a wrong assumption here?

    Best regards
    Caudio

     

    0
  • Great, Thanks a lot for your feedback

    Claudio

    0

Please sign in to leave a comment.