Forecasting client retention — A machine-learning approach

Abstract In the age of big data, companies store practically all data on any client transaction. Making use of this data is commonly done with machine-learning techniques so as to turn it into information that can be used to drive business decisions. Our interest lies in using data on prepaid unitary services in a business-to-business setting to forecast client retention: whether a particular client is at risk of being lost before they cease being clients. The purpose of such a forecast is to provide the company with an opportunity to reach out to such clients as an effort to ensure their retention. We work with monthly records of client transactions: each client is represented as a series of purchases and consumptions. We vary (1) the length of the time period used to make the forecast, (2) the length of a period of inactivity after which a client is assumed to be lost, and (3) how far in advance the forecast is made. Our experimental work finds that current machine-learning techniques able to adequately predict, well in advance, which clients will be lost. This knowledge permits a company to focus marketing efforts on such clients as early as three months in advance.

[1]  Jonathan Loo,et al.  Customer churn prediction in telecommunication industry using data certainty , 2019, Journal of Business Research.

[2]  Dirk Van den Poel,et al.  Predicting customer retention and profitability by using random forests and regression forests techniques , 2005, Expert Syst. Appl..

[3]  João Falcão e Cunha,et al.  Customer attrition in retailing: An application of Multivariate Adaptive Regression Splines , 2013, Expert Syst. Appl..

[4]  Koen W. De Bock,et al.  An empirical evaluation of rotation-based ensemble classifiers for customer churn prediction , 2011, Expert Syst. Appl..

[5]  Peter E. Hart,et al.  Nearest neighbor pattern classification , 1967, IEEE Trans. Inf. Theory.

[6]  Michel Ballings,et al.  Customer event history for churn prediction: How long is long enough? , 2012, Expert Syst. Appl..

[7]  Arno De Caigny,et al.  A new hybrid classification algorithm for customer churn prediction based on logistic regression and decision trees , 2018, Eur. J. Oper. Res..

[8]  Tammo H. A. Bijmolt,et al.  Staying Power of Churn Prediction Models , 2010 .

[9]  Yoav Freund,et al.  A decision-theoretic generalization of on-line learning and an application to boosting , 1997, EuroCOLT.

[10]  Li Xiu,et al.  Application of data mining techniques in customer relationship management: A literature review and classification , 2009, Expert Syst. Appl..

[11]  Nima Jafari Navimipour,et al.  Customer relationship management mechanisms: A systematic review of the state of the art literature and recommendations for future research , 2016, Comput. Hum. Behav..

[12]  Ali Tamaddoni Jahromi,et al.  Managing B2B customer churn, retention and profitability , 2014 .

[13]  Wei-Yin Loh,et al.  Classification and regression trees , 2011, WIREs Data Mining Knowl. Discov..

[14]  Markus Haltmeier,et al.  A machine learning framework for customer purchase prediction in the non-contractual setting , 2020, Eur. J. Oper. Res..

[15]  J. Sheth,et al.  Customer Relationship Management: Emerging Practice, Process, and Discipline , 2002 .

[16]  Özden Gür Ali,et al.  Dynamic churn prediction framework with more effective use of rare event data: The case of private banking , 2014, Expert Syst. Appl..

[17]  Bart Baesens,et al.  New insights into churn prediction in the telecommunication sector: A profit driven data mining approach , 2012, Eur. J. Oper. Res..

[18]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[19]  Ronald Swift,et al.  Accelerating Customer Relationships: Using Crm and Relationship Technologies , 2000 .

[20]  Kristof Coussement,et al.  Faculteit Economie En Bedrijfskunde Hoveniersberg 24 B-9000 Gent Churn Prediction in Subscription Services: an Application of Support Vector Machines While Comparing Two Parameter-selection Techniques Churn Prediction in Subscription Services: an Application of Support Vector Machines While Comparin , 2022 .

[21]  Chih-Fong Tsai,et al.  Customer churn prediction by hybrid neural networks , 2009, Expert Syst. Appl..

[22]  David G. Kleinbaum,et al.  Logistic Regression. A Self- Learning Text , 1994 .

[23]  Qiang Yang,et al.  Telco Churn Prediction with Big Data , 2015, SIGMOD Conference.

[24]  Koen W. De Bock,et al.  Reconciling performance and interpretability in customer churn prediction using ensemble learning based on generalized additive models , 2012, Expert Syst. Appl..

[25]  David C. Yen,et al.  Customer Relationship Management: An Analysis Framework and Implementation Strategies , 2001, J. Comput. Inf. Syst..

[26]  D. Quinn Mills,et al.  Customer Management as the Origin of Collaborative Customer Relationship Management , 2004 .

[27]  Stephen R. Marsland,et al.  Machine Learning - An Algorithmic Perspective , 2009, Chapman and Hall / CRC machine learning and pattern recognition series.

[28]  Minghe Sun,et al.  A hierarchical multiple kernel support vector machine for customer churn prediction using longitudinal behavioral data , 2012, Eur. J. Oper. Res..

[29]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[30]  Kristof Coussement,et al.  Improved marketing decision making in a customer churn prediction context using generalized additive models , 2010, Expert Syst. Appl..

[31]  Sahar F. Sabbeh,et al.  Machine-Learning Techniques for Customer Retention: A Comparative Study , 2018 .