Why does the 0-1 bucket in 'total events per user' behavior analysis show no users?
Why the 0-1 bucket in Total Events per User behavior analysis is always empty, and how to find users who never performed an event by using segmentation.
When you run a Total Events per User behavior analysis with a custom distribution, the 0-1 bucket always shows 0 users, even though you expect it to capture low-activity users.
Two things combine to make this bucket permanently empty:
Total Events per User counts only users who performed the selected event at least once. The analysis measures how many times each user performed an event they have already performed, so a count of 0 can never appear in the result set.
Custom distribution buckets exclude their upper bound. A 0-1 bucket captures values greater than or equal to 0 and less than 1 — that is, only a count of exactly 0. A user with exactly 1 event falls into the next bucket, 1-2, not 0-1.
Because no user in this analysis can have a count of 0, the 0-1 bucket is always empty, regardless of your data.
To see a clean distribution of actual usage frequency, set Distribution bucketing to Custom and set From to 1 instead of 0. The buckets are generated from the From, to, and interval values, so starting at 0 always produces an empty first bucket.
Behavior analysis cannot surface users who never performed an event, because it analyzes only the users who did. To find and act on that zero-event cohort, use segmentation instead:
Go to Segment > Create segment and select Filter users by.
On the User behavior tab, select Has Not Executed and choose your event — for example, App/Site Opened.
Set the timeframe you want, for example in the last 7 days.
Click Show count to check the size of the cohort, then save the segment.
Use this segment directly for analysis, or to build a re-engagement campaign that targets users who have not performed the event.
Assistant
Responses are generated using AI and may contain mistakes.
