Every video service we work with has a fairly substantial number of subscribers that haven’t watched a video in the last 30 days. We call these inactives. Intuitively, these customers are at-risk of leaving the service and this is validated by our machine learning application, the Customer Happiness Index. A lack of activity generally has a strong causal relationship with churn.
So when do you reach out to inactives to drive them back to your video service? Too soon or too frequently and it may be annoying. Too late and you might accidentally drive churn higher as you remind users they have a membership they are no longer using. The following is a methodology for finding the sweet spot for re-engaging your inactive audience members.
Creating Your Cohorts
The first step is to develop your target lists. With the Subscriber Export built into the Wicket Scorecard, this is straightforward. On the Customers wicket, click on the download icon next to Net (Fig. 1). There’s no need to apply a filter as we’ll start with all subscribers.
With the subscriber list downloaded, sort based on lastViewDate to create a set of cohorts. We recommend grouping your inactives into 10 days, 11 to 20 days, 21 to 30 days, 31 to 45 days, 46 to 60 day and 61 to 90 days since they last watched a video. For new or small video services you may include all inactive customers. For larger services, taking a sample of 600 inactives in each cohort should be representative and easier to work with later. You may also want to remove those customers on a one-year or two-year subscription plan as they may skew the churn numbers. As a last step, divide each cohort in half, tagging an A and a B group.
Next up, develop an email to reach out to the B’s. This can be a general message promoting new or upcoming shows, movies or events as well as other content the audience may enjoy. Load up the B’s into your marketing or email system using the appropriate hashed email field and hit send.
Calculating the Results
With the email away, we recommend waiting a week and then running subsequent reports for the next three weeks to give you a solid month of results. This is especially important for tracking any up-ticks in churn.
Calculating the results takes a little effort but is generally straightforward. The first step is to run the same data export as before. Next up is to match the new list with your starting list, taking care to keep the A’s and B’s for each cohort labeled. From this, we’re looking for differences in two fields, LastViewDate and ActiveEndDate. A simple subtraction between the same fields on different dates will illuminate who has come back to watch a video and those subscribers that have chosen to leave the service. Converting each delta to a 1 makes summing and calculating percentages easy.
As a last step, a pivot table may quickly answer the following questions:
For each cohort
- What percent of the A’s came back to watch a video?
- What percent of the B’s came back to watch a video?
- What percent of the A’s unsubscribed for any reason?
- What percent of the B’s unsubscribed for any reason?
Assess the results
Our hypothesis is that every video service has an optimal window for outreach. Between 10 and 30 days may drive the best re-engagement with the service but this window may extend to 45 or even 60 days. Lost customers may creep up for the B’s but a little additional churn for the younger cohorts will likely be outmatched by the value of re-engaging a larger number of inactives. At some point, this will flip and the increased churn will offset the value of the re-engagement.
Finding the sweet spot for your inactives takes a little effort but the payoff is reduced churn and a growing Audience Lifetime Value. Once you have this knowledge you can operationalize your outreach by designating a team member to send a weekly email or notification to the inactives in the optimal window. If you’d like to learn more about using the Wicket Scorecard and its actionable insights to run these tests and target inactive customers, please contact us to set up a demo and conversation.
Tags: lost customers • Subscriber export • subscriptions