Non-Monetary Calculations for Customer Lifetime Value Model

Christian Thøgersen -

In the other CLV Model guides, it has largely been treated as a given that the models are applied to purchases and orders. In most situations, this is indeed what you will want to do, using either the Price & Quantity or Subtotal Churn Schemas. However, contrary to popular belief, there does exists values that cannot be measured in dollars and cents… such as customer loyalty or eyes-on-page metrics, for instance. Fortunately, the CLV Model is quite capable of modeling this too! Selecting the Other events Churn Schema when you create the CLV Model will switch the Attributes to this setup, at which point all of the Attributes you usually use to measure purchases and subtotals take on an alternate meaning relating to visits and page-views instead. The rest of the CLV Model will function just as described in the regular How-To guide for the feature.

 

List of Customer Lifetime Value attributes for Event Values

 

  • Predicted alive (%): Predicted alive represents the probability of the customer visiting a product-page again at some point in the future. It is otherwise calculated as normal, and can be seen as the counterpart to ‘attention-churn’.
  • Historic value last 365 days: The customer’s total number of page-views for the last year.
  • Historic value last 365 days: The customer’s total number of page-views for all time – or rather, going back as far as the dataset itself.
  • Predicted future value next 365 days: The customer’s expected number of page-views for the next year.
  • Predicted Customer Lifetime Value: The expected number of page-views for the entire lifetime of the customer, adding together the historic value for the length of the dataset and a long-term prognosis.
  • Predicted number of orders next 365 days: Expected number of page-views for the next year. Using this form of value, it becomes effectively identical to Predicted future value.
  • Days since first order: Number of days since the customer’s first visit to your website.
  • Days since last order: Number of days since the customer’s latest visit to your website.
  • Number of orders: Number of discreet days during which the customer has visited your website at least once. Multiple visits on the same day still only count as one.
  • Average order value: The customer’s average number of page-views per visit.
  • Average days between orders: The average number of days between the customer’s visits to your site.
  • Inactivity score: This measures how close the customer currently is to their average frequency of visiting your site. Just as when it is applied to purchases, a score of around 100 suggests that they’re right on schedule, a score below that point means that their most recent visit came sooner than average, while a score significantly above 100 might mean that they have ‘lapsed’ and gone elsewhere – or at least visit your page rather more rarely as of late.

 

Hopefully, this will clear up any confusion when it comes to applying the CLV Model to values other than purchases.

Have more questions? Submit a request

Comments

Powered by Zendesk