The client is an India's largest telecom operator with its headquarter based in Mumbai, Maharashtra. It is a pan-India integrated GSM operator offering 2G, 3G and 4G mobile services under two brands named Vodafone and Idea.


"Facing customer attrition issue Wanted to identify the customers who were likely to attrite so that retention marketing campaigns could be executed Wanted to do a pilot before nation wide rollout "


"Xerago collaborated with the network provider, reviewed the marketing challenge and identified an analytics led churn model as the solution.A churn model uses predictive analytics to identify customer behavioral patterns that indicate likelihood of customer attrition. Used Unica Affinium 5.8 to build a churn model and identified the top three segments that were likely to churn. Started by defining “Attrition” profile. Identified that a logistic model using customer usage data on Voice Calls, SMS, VAS, Roaming and Recharges made would be appropriate. Performed analysis and modeling on the remaining customer base by deriving over 100 different variables using the input variables. Used CHAID Algorithm to predict the churn behavior and identified the top three segments that were likely to attrite "


"Xerago built a churn model that could predict customer attrition with 71% accuracy.The results of this model were used to run outbound retention campaigns that dramatically reduced churn rate and increased retention.The success of this assignment formed the business case for nationwide churn model to identify customer attrition. "

Get in Touch

For queries regarding our Customer Cloud or our Professional Services

* Mandatory

captchaLoading Page - Xerago

By submitting this form, I provide consent to be contacted by Xerago for promotions. Please select your preferred communication mode:


Xerago takes your privacy very seriously. Please see our Privacy policy for more information on how we use your data.



Thank You

for reaching out to us. One of our representatives will get in touch with you shortly.