1. QBE Overview & Available Data
As a mature organization, QBE assists its customers to manage their online presence through a subscription service with contracts that ran either month to month or fixed periods of 6 months or 12 months. Moreover, QBE collects various pieces of information focus to further understand its customers, nature of this study which seeks to predict the probability for a customer to cancel their service. I will attempt to recognize the patterns and behaviors that expose the factors needed to be improved in order to increase QBE’s customer retention levels.
The available information to prepare the analysis is concentrated in the months of November, December, and the changes between these two periods: 2. Analysis & …show more content…
QBE dependence of churn rates on customer age
In analyzing the dependency of churn in a relationship to the customer age, the x-axis identifies the number of months with service and the y-axis represents the probability to churn based on the customer longevity. Contrary to the management theory, it can be observed that as the age increases the probability to churn also increases. Perhaps the loyalty is not a determining indicator of the customer wiliness to continue with the services.
5. Customer 672 probability to leave (High or Low …show more content…
Also, in the following statistical equation, the # 148 represents the CHI score for their current customer ID 672. As indicated in the equation, the probability of customer 672 is equal to 3.14% which approximately represent less than 1/3 of the highest probability of 9.85 for customer 51.
6. Analysis of QBE single model approach
In analyzing the QBE single model approach, the most interesting relationship that I was able to expose was CHI score for the month of December versus the Churn factor. The lesson learned indicates that as the CHI score increases the probability for a customer to cancel is less likely to happen. Certainly, this model does not represent a clear relationship between other factors such as customer age or blogs which in most cases represent an opposite behavior.
7. QBE best-contributing factors to the predictive
As a mature organization, QBE assists its customers to manage their online presence through a subscription service with contracts that ran either month to month or fixed periods of 6 months or 12 months. Moreover, QBE collects various pieces of information focus to further understand its customers, nature of this study which seeks to predict the probability for a customer to cancel their service. I will attempt to recognize the patterns and behaviors that expose the factors needed to be improved in order to increase QBE’s customer retention levels.
The available information to prepare the analysis is concentrated in the months of November, December, and the changes between these two periods: 2. Analysis & …show more content…
QBE dependence of churn rates on customer age
In analyzing the dependency of churn in a relationship to the customer age, the x-axis identifies the number of months with service and the y-axis represents the probability to churn based on the customer longevity. Contrary to the management theory, it can be observed that as the age increases the probability to churn also increases. Perhaps the loyalty is not a determining indicator of the customer wiliness to continue with the services.
5. Customer 672 probability to leave (High or Low …show more content…
Also, in the following statistical equation, the # 148 represents the CHI score for their current customer ID 672. As indicated in the equation, the probability of customer 672 is equal to 3.14% which approximately represent less than 1/3 of the highest probability of 9.85 for customer 51.
6. Analysis of QBE single model approach
In analyzing the QBE single model approach, the most interesting relationship that I was able to expose was CHI score for the month of December versus the Churn factor. The lesson learned indicates that as the CHI score increases the probability for a customer to cancel is less likely to happen. Certainly, this model does not represent a clear relationship between other factors such as customer age or blogs which in most cases represent an opposite behavior.
7. QBE best-contributing factors to the predictive