Churn risk scoring using call network analysis
Customer churn risk scores are based on a multi-variable churn risk model relating customer and customer account characteristics to a risk of customer churn. A computer-implemented method of generating and presenting churn risk scores of customers of a telecommunication provider involves analyzing, on an in-memory database platform, customer call data records and customer records to calculate a churn likelihood value, an influence factor value, and an average spend value for each customer. The method assigns a churn risk score to each customer according to the model using the calculated churn likelihood value, the calculated influence factor value, and the calculated average spend value as input to the model. The churn risk scores for one or more customers are displayed visually on an interactive computer-user interface (UI).
1. A computer-implemented method comprising:
presenting an interactive computer-user interface (UI) of a web-based churn risk analytics application,
the interactive UI including a web page with user controls of operations of the web-based churn risk analytics application;
analyzing, by the web-based churn risk analytics application, on an in-memory database platform, multiple data sources including customer call data records and customer records related to call networks of a telecommunication provider to calculate a churn likelihood value, an influence factor value, and an average spend value for each customer;
assigning multiple churn risk scores to each customer according to a multi-variable churn risk model relating customer and customer account characteristics to a risk of customer churn, the multi-variable churn risk model using, as inputs, the calculated churn likelihood value, the calculated influence factor value, and the calculated average spend value and producing, as outputs, the multiple churn risk scores;
displaying the multiple churn risk scores and churn risk calculations as a three dimensional visualization on the interactive UI to identify at risk customers in real time or near real time; and
providing drill-down support on the interactive UI, the drill-down support including a churn breakdown graph depicting connections between a customer and other users and an influence breakdown graph depicting incoming and outgoing calls between a customer and other users.
2. The method of claim 1 , wherein displaying the multiple churn risk scores and churn risk calculations as a three dimensional visualization includes displaying a calculation view on the UI and consuming the calculation view in analyzing customer call data records and customer records via Open Data Protocol (OData).
3. The method of claim 1 , wherein displaying the multiple churn risk scores and churn risk calculations as a three dimensional visualization includes displaying the churn risk scores in a 3-dimensional graph with the churn likelihood, the influence factor, and the average spend variables as the graph axes.
4. The method of claim 3 , wherein displaying the multiple churn risk scores and churn risk calculations as a three dimensional visualization for one or more customers visually on an interactive computer user interface includes displaying an ordered list of at-risk customers or customer accounts.
5. The method of claim 3 , wherein displaying the multiple churn risk scores and churn risk calculations as a three dimensional visualization includes configuring UI operations to allow a viewer to explore the displayed customer data and the underlying model calculations.
6. The method of claim 3 , wherein displaying the multiple churn risk scores and churn risk calculations as a three dimensional visualization includes providing drill down support on the UI to show individual graphs depicting individual call networks of the one or more customers.
7. The method of claim 1 , wherein analyzing customer call data records and customer records database includes using social network analysis techniques to extract the one or more customers' social call networks.
8. The method of claim 7 further comprising displaying the one or more customers' social call networks on the UI in graphs.
9. The method of claim 7 further comprising, using social network analysis measures including at least a connectedness measure to characterize the one or more customers' social call networks.
10. The method of claim 9 further comprising using social network analysis measures to calculate the influence factor value.
11. A computer system comprising a memory and a semiconductor-based processor, the memory and the processor forming one or more logic circuits configured to at least:
present an interactive computer-user interface (UI) of a web-based churn risk analytics application,
the interactive UI including a web page with user controls of operations of the web-based churn risk analytics application;
analyze, by the web-based churn risk analytics application, multiple data sources including customer call data records and customer records related to call networks of a telecommunication provider to calculate a churn likelihood value, an influence factor value, and an average spend value for each customer;
assign multiple churn risk scores to each customer according to a multi-variable churn risk model relating customer and customer account characteristics to a risk of customer churn, the multi-variable churn risk model using, as inputs, the calculated churn likelihood value, the calculated influence factor value, and the calculated average spend value and producing, as outputs, the multiple churn risk scores;
display the multiple churn risk scores and churn risk calculations as a three dimensional visualization on the interactive UI to identify at risk customers in real time or near real time; and
provide drill-down support on the interactive UI, the drill-down support including a churn breakdown graph depicting connections between a customer and other users and an influence breakdown graph depicting incoming and outgoing calls between a customer and other users.
12. The computer system of claim 11 , wherein the logic circuits are configured to display a calculation view on the UI and consuming the calculation view in analyzing customer call data records and customer records via Open Data Protocol (OData).
13. The computer system of claim 11 , wherein the logic circuits are configured to display the churn risk scores in a 3-dimensional graph on the UI with the churn likelihood, the influence factor, and the average spend variables as the graph axes.
14. The computer system of claim 11 , wherein the logic circuits are configured to display an ordered list of at-risk customers or customer accounts.
15. The computer system of claim 11 , wherein the logic circuits are configured to enable UI operations to allow a viewer to explore displayed churn risk scores and the underlying model calculations.
16. The computer system of claim 11 , wherein the logic circuits are further configured to provide drill down support on the UI to show individual graphs depicting individual call networks of the one or more customers.
17. The computer system of claim 11 , wherein the logic circuits are further configured to analyze customer call data records and customer records database by using social network analysis techniques to extract the one or more customers' social call networks.
18. The computer system of claim 17 , wherein the logic circuits are further configured to display the one or more customers' social call networks on the UI in graphs.
19. A non-transitory computer readable storage medium having instructions stored thereon, including instructions which, when executed by a microprocessor, cause a computer system to at least:
present an interactive computer-user interface (UI) of a web-based churn risk analytics application,
the interactive UI including a web page with user controls of operations of the web-based churn risk analytics application;
analyze, by the web-based churn risk analytics application, multiple data sources including customer call data records and customer records related to call networks of a telecommunication provider to calculate a churn likelihood value, an influence factor value, and an average spend value for each customer;
assign multiple churn risk scores to each customer according to a multi-variable churn risk model relating customer and customer account characteristics to a risk of customer churn, the multi-variable churn risk model using, as inputs, the calculated churn likelihood value, the calculated influence factor value, and the calculated average spend value and producing, as outputs, the multiple churn risk scores; and
display the multiple churn risk scores and churn risk calculations as a three dimensional visualization on the interactive UI to identify at risk customers in real time or near real time; and
provide drill-down support on the interactive UI, the drill-down support including a churn breakdown graph depicting connections between a customer and other users and an influence breakdown graph depicting incoming and outgoing calls between a customer and other users.
20. The non-transitory computer readable storage medium of claim 19 , wherein the instructions, when executed by a microprocessor, further cause the computer system to:
display the churn risk scores in a 3-dimensional graph with the churn likelihood, the influence factor, and the average spend variables as the graph axes.