Cohorts Overview

Cohorts are groups of users defined by a chosen set of criteria, like a shared property or sequence of events. When you define a cohort, you can use it to group and filter data in the Analysis reports. You can also use cohorts to target groups of users via messaging integrations. 

Use Cohorts in Reports

Cohorts can be created almost anywhere within Mixpanel.  

You can create a new cohort from the Insights, Flows, Funnels, Retention, and Impact Reports. To do so, click "Create Custom" > "Cohort" from within any Filter or Breakdown menu.

mceclip0.png

You can also create cohorts directly from the charts in Retention reports or Funnels reports.

You can create and save cohorts in the Users page as well.

Finally, cohorts can be created via the Create Cohort button at the top right of the Cohorts management page, found in the Data Management menu.

After you create a cohort, you can use it to analyze your user data in the Analysis Reports.

Visualize Cohorts Over Time

Select a cohort from the Insights drop down menu to see how the number of users in the cohort has changed over time.

angela-gif.gif

When you visualize a cohort over time, you can breakdown by another cohort, compare to an event, or compare to another cohort.

Note on cohorts whose filters contain user profile properties:

Because user profile properties only store the most recent value, cohorts involving user profile properties will use the current value for those properties (even if the value changed over time).

For example, suppose the cohort's filter criteria is: users where user[“City”] == “SF” and “Did Event: Order Ride 3 times in Last 7 days”.

Mixpanel computes the above over the last 30 days and then groups the set of users who have property user[“City”] == “SF” as of right now and intersect that with the daily cohort of users who did Order Ride 3 times in the last 7 days.

Cohorts Characteristics

Cohorts have several characteristics that affect results in Mixpanel reports. While Cohorts are intuitive, the following traits are important to realize when interpreting results.

Cohorts are Dynamic

Users qualify for a cohort based on the specified conditions. This doesn’t ensure that that user will always be in the cohort.

For example, assume the condition of the cohort is performing a “Song Play” event in the last 30 days. If a user plays a song once in thirty days, they will enter the cohort. If that user doesn’t listen to another song at the 30 day mark, however, then they will exit the cohort as they no longer meet the qualifying condition of having performed a “Song Play” event in the last 30 days.

Cohorts are Always Current at the Time you Run a Report

When you run a report that involves a cohort, Mixpanel returns the group of users that meet the criteria for that cohort at the moment.

This is important to consider when running a report with a “From Date” and “To Date” selected. The analyzed cohort will not be the cohort during the “From Date” and “To Date” time period. The analyzed cohort will be users currently in the cohort.

Export Cohorts

From Data Management -> Cohorts, click Export in the inline action menu to export a cohort list to a webhook or an integration 

Untitled.png

To download a csv, click View Users and then click Export CSV. The CSV file will automatically download. 

Note:

  • The export will only show the user profile properties of users in the cohort that are viewable in the currently displayed table at the time of export. You don't get ALL of a user's profile properties.
  • Users who exist in the cohort that don't have a user profile will appear in the CSV as a distinct_id only.

Delete Cohorts

Click on the trash icon or click Delete in the inline action menu to delete a cohort.

Untitled2.png

If you create cohorts based off of other cohorts, you will not be able to delete the original cohort the new cohort is dependent on.

For example, you can create a new cohort A which combines cohorts B and C, but you will not be able to delete cohorts B or C, until you delete cohort A.

Did this answer your question?

Comments

0 comments

Article is closed for comments.