Reduced customer churn in a leading gaming corporation

  • June 11th, 2017

What We Did: Established a robust framework for early detection of players likely to churn, along with improvements in marketing strategy for customer retention.

The Impact We Made: Improvements in design of marketing campaigns for customer retention led to improved ROI. Churn rate for players was reduced by 14% from the earlier rate of 46%.

Summary – Identifying reasons for customer churn

The client is a leading player in the hospitality and entertainment sector. They had pre-defined rules/ logic to identify likely decliners in engagement. These decliners were then targeted through specific marketing campaigns to prevent churn. This process, established to tackle revenue loss due to withering engagement with customers, was unable to identify reasons of customer churn.

The client wanted to establish a robust framework which helps in:

  • Being precise in selecting the target population (likely decliners in engagement)
  • Identifying the target population as early as possible
  • Knowing which marketing action will have the greatest retention impact on each customer segment

About The Client – Large gaming corporation

The client is one of the leading gaming corporation in the world with annual revenues of over $10 Bn. It operates over 50 casinos and hotels worldwide.

The Challenge – Tackling revenue loss due to withering consumer engagement

The existing process or framework was unable to identify and target population that was churning or declining; it lacked precision. Also, the defined rules for decliner identification were based on heuristic information rather than being data driven.

The Approach: Automation to support effective retention strategies

Mu Sigma team improved the current retention strategy of the client by:

  • Studying the behavioral patterns exhibited by the players which signal the risk and timing of customer churn
  • Building a logistic regression model that leveraged the information to determine the propensity of a customer to decline
  • Customer life-time value was calculated for each of the player and incorporated into the propensity score
  • Customers/ players were then profiled on the basis of their propensity score and a reinvestment/ promotion effort was performed in accordance

Outcome – Improved churn management and marketing campaign

The improved churn mechanism helped the client make better decisions and organize relevant marketing campaigns:

  • The client is now able to use the propensity score from the model to prioritize customers for various marketing campaigns and increase engagement. This led to an improvement in ROI on marketing spend by 11%.
  • Early detection of churn enabled the client to reactivate declining players with attractive offers. The churn rate of players was reduced to 32% from 46%.