Segment app users based on their usage frequency
I'm struggling a bit to find the right approach of a segmentation:
We are tracking an app and I would like to build clusters based on the users weekly usage of the app, e.g.
Occasional visitors: <=1 visits/app launches per week
Frequent visitors: >1 and <=5 visits/app launches per week
Heavy users: >5 visits/app launches per week
I would later like to visualize their share over time in a 100% stacked bar chart.
How would you recommend to define these segments? Please note, that I do not have "Return Frequency" available. I guess this is due to tracking and app and not a website, but not sure. It's just not available in the workspace (no restrictions set). Instead I have
- Metric: App Users (Unique Visitors)
- Metric: App Launches (Higher than visits. Multiple launches within 30 minutes timeout count as one visit)
- Metric: Visits
- Dimension: Days since first use
- Dimension: Days since last use
The two dimensions could be bucketed using classification.
How would you recommend to define the 3 segments?


