How to setup power user curve in google analytics?

[CarlosDubois]

Hey all,

Has anyone implemented a power user curve in Google Analytics?

I tried to do it for our product, however, I couldn’t figure it out. The cohort analysis comes closest, however, cohort based on acquisition date doesn’t work for our product. We’re interested in usage frequency of returning users on a continuous monthly basis.
For reference: The Power User Curve: The Best Way to Understand Your Most Engaged Users - Andreessen Horowitz

[RohitKumar]

I have been doing power user and frequency of usage analysis so often, and not even once I used GA for it. GA is focused on user acquisition, but for this analysis, you rather have to look at your active users, which you define based on your product usage (not sure if you send this activity data to GA in the first place). My recommendation would be:

1. Define Active users definition and segment
2. The most correct way would be to run this analysis in SQL if you store visits/activity.
3. If you don’t, you also can do this in Looker/ Amplitude/ Mixpanel/ any other web analytics tool with some limitations. Would have to create a custom event for activity and play around with cohorts.

[CarlosDubois]

Thanks @RohitKumar , we’re doing two things now:

1. Use the G-Analytics “User Exploration” report to identify the power users based on duration on the platform. We are then creating segments to compare their behavior to others.
2. We are tapping the data directly and using SQL to add the power curve to our Data Studio dashboard.

[NatalieSmith]

Hi @CarlosDubois @RohitKumar , I am trying to better understand the Power-user Curve and don’t quite understand the right side of the smile increasing around days 26-30. Why would engagement or MAUs increase on these latter days of the month? I am certain that I am misunderstanding something and appreciate any insights you can share.

[RohitKumar]

It doesn’t have to increase, but it is often the case for successful products (features). It means users use the product on average 26-30 days per month or, in other words, every day. This is very good. If the chart is skewed towards the left, it says users try out the product and churn. Or use it only a few times per month. Skewed to the right indicates that users are highly engaged.