top of page

Case study

Customer churn

Finding the right treatment for the right customer at the right time.

Challenge

Solution

  • Ensuring high returns on marketing investments is essential; however, retention campaigns can be costly and carry inherent risk. A poorly executed campaign can even itself cause customer churn.


  • Randomized A/B tests don’t understand confounding factors and offer limited customization needed to target distinct customer segments effectively.


Impact

  • A Causal AI is trained to identify the factors affecting churn of each customer. 


  • Based on this model, campaign simulations are run on subsets of customers instead of a “one-size-fits all” intervention. This level of personalization can improve effectiveness and significantly reduce risk.


  • Explainable recommendations are given to marketing teams to plan individual promotional strategies. Cost-efficiency is ensured by directing resources towards customers who are genuinely at risk of churning.

  • Reduction in customer churn, leading to an increased CRR.


  • Effective distribution of marketing resources, lowering CRC.


  • Insights into customer behaviors and promotional campaign efficacy.

bottom of page