Skip to content
ML Academy

Experimentation and causality

Interview question

  • seniorlevel
  • 12 minsuggested time
  • 2Data scientist, ML engineer

A telecom company sends retention discounts to the 10% of customers with the highest churn-model scores. A stakeholder says: "That's obviously the best targeting." Do you agree? How would you decide whom to target?

Tests: Distinguishing prediction (churn risk) from causal effect of an intervention (uplift), and designing experiments to estimate heterogeneous treatment effects.

day 1day 5day 13

Sign in to practise: answers are saved before model answers are shown.