Build an enagement matrix. This matrix will count users, it will count events, it will show the % of daily active users and the average times that users perform events. Once we split by event type we can see what % of users perform that event, and the average number of times they perform it.
Build a query in Explore using table view. This query will require 4 Measures. Measure 1 should count unique user actors filtered to all user actors as a fraction of all. Measure 2 should count of events filtered to all events. Measure 3 should count unique user actors filtered to all user actors. Measure 4 should divide measure 2 by measure 3. Splity by action. The query should run from the beginning of time to now. In chart options you can change Measure 1 and 4 to %.
An engagement matrix is a very powerful way to determine what buttons, features or actions your users are using, how often they're doing those actions and the percent of users who are doing those actions on a daily basis. This is a core foundation of determining feature stickiness, especially if you compare it to segments of highly engaged and low engaged users.
Note: The numbers and result does not need to be an exact match as the data changes over time. The query construction is what is most important. See images below.
*Click to enlarge*