The Stickiness metric measures the number of days a user performs an event in a weekly or monthly interval so that you can understand which events are driving users back to your product.
Table of Contents
Stickiness will help you learn about the distribution of engagement levels across your user base and how many users have formed product usage habits. This metric can be measured in two ways: Cumulative and Non-cumulative. Cumulative measures the percentage of users for each interval who were at that level of stickiness or greater. Non-cumulative measures the percentage of users for each interval who were at that level of stickiness exactly.
The breakdown of data is based on days after each cohort's start date. For example, "2 Days" means that the user has done the event on two or more days that week (or month). "3 Days" means that the user has done the event on three or more days that week (or month).
The bar graph averages all of the Nth-Day stickiness numbers from the user cohorts within the selected timeframe. By default, the graph will show the stickiness of users who have performed any event. You can specify the event in the left panel of the chart controls.
You can also click on a specific data point to inspect the users that make up that point. See Microscope for more information.
Non-cumulative stickiness, calculates the percentage of users-weeks (or user-months if the interval is a month) that were active for exactly X number of days in the week or month. This provides a more precise measure than cumulative.
The breakdown of data is based on days after each cohort's start date. For example, "2 Days" means that the user has done the event on exactly two days that week (or month). "3 Days" means that the user has done the event on exactly three days that week (or month).
The chart will display the percentage of user-weeks that were active during your selected timeframe. If a user was 1-day sticky for Week 1 and then 2-day sticky for Week 2, that user would contribute a user-week to 1-day stickiness and a user-week to 2-day stickiness. X-day stickiness is computed as the number of user-weeks that were x-day sticky over the total number of user-weeks.
The table shows a detailed breakdown of the data by each user cohort and more granular day buckets. In the example below, there were 290,149 users in the week beginning with October 17th. The Day 2 Cumulative stickiness is 76.8%, meaning that 222,834 out of the 290,149 users did an event on two days or more that week. Days with incomplete data will have an asterisk.
Stickiness Over Time
Stickiness Over Time shows stickiness rates over time. When looking at the Weekly mode, you can see how your 2-day, 3 -day, 5-day, and 7-day stickiness change over time. Additionally, in the monthly Stickiness mode, you are able to see how your 3-day, 9-day, 15-day, 21-day, and 30-day stickiness change over time.