7 Segmenting customers with advanced metrics
This chapter covers
- Metrics made from ratios of other metrics
- Metrics that measure behavior as a percent of a total
- Metrics that show how behavior changes over time
- Measuring metrics with long periods and presenting them as metrics with shorter periods, and visa versa
- Metrics for multi-user accounts
You’ve learned a lot about understanding churn with metrics derived from events and subscriptions. You've seen that simple behavioral measurements can be very powerful at segmenting customers who may be at risk for churn, and have different levels of engagement. But you've also seen some of the limitations of simple behavioral metrics: Usually many simple metrics are correlated with each other. Correlations arise because customers who have a lot of any one-product--related events tend to have a lot of the other events as well. And the correlations makes it hard to tell which types of behaviors are actually most important.
But the problem is actually deeper than a lack of refinement. In this chapter you'll learn that correlation between metrics can actually make you misread the influence of a behavior. A behavior that’s negative in the sense that it takes utility and enjoyment away from customers may actually appear to enhance engagement when correlated with other behaviors that provide utility and enjoyment.