Step 4: Segment the data
The task below will be automatically checked off once you complete watching the video.
Now that we’ve seen which metrics are predictive and which are not, focus on the predictive metrics to target members who are not likely to renew. To do this, we’ll need to create segments/cohorts based on the metrics.
First, take each predictive metric and sort the members for whom you know the outcome (that is, whether they renewed in a given timeframe or not) from low to high. For example, if you identify CEUs as a predictive metric where the more someone attains CEUs, the less likely they are to leave, sort your members from the least CEUs attained to the most.
Split the list into 10 segments, or cohorts. For each cohort, calculate how many members left the organization to find your churn rate for each cohort. When you plot it out, you’ll see the lowest decile has the highest churn rate.
Note: Target U-shaped curves differently. Newer members warrant a different communication strategy than tenured members.
Tip: Use percentiles because it shows a nearly equal number of members per decile allowing you to have more direct comparisons.
Open or download the Predicting Member Churn Spreadsheet to see the examples of segments by viewing the cohorts tabs in the spreadsheet.
[accessally_course_navigation prev_button='Previous' next_button='Next']
[progressally_note note_id='1' allow_attachment='no']
Share your progress!
[progressally_social_share type='facebook' share_id='1' post_id='42'][/progressally_social_share][progressally_social_share type='twitter' share_id='1' post_id='42'][/progressally_social_share][progressally_social_share type='email' share_id='1' post_id='42'][/progressally_social_share]
[progressally_progress_bar post_id="6437" width="100%" height="25"]