Database
Marketing & Customer
strategy Management Vol. 14, 2, 130–142 © 2007 Palgrave Macmillan Ltd 1741-2439 $30.00www.palgrave-
留学生论文代写journals.com/dbmAndreas HerrmannCenter for Business MetricsUniversity of St GallenGuisanstrasse 1a,St. Gallen 9000, SwitzerlandTel: + 41 71 224 21 31;Fax: + 41 71 224 21 32;e-mail:
[email protected]OutlineAccording to the literature, investing incustomer retention is often more effi cientthan acquiring new customers. 1 Hence, aconsiderable amount of research hasaddressed factors and strategies to retaincustomers. several papers, however, —including Anderson and Mittal, 2 Nirajet al ., 3 and Reinartz and Kumar 4 —emphasise that it may not be profi table toretain every customer. Rather, therelationship between costs and value mustbe taken into consideration; that is,customers are not profi table if the costsgenerated to serve them exceed revenuesin the long term. Thus, fi rms shouldconsider reducing services for thesecustomers or even terminate the relationshipentirely. 5The benefi ts of customer segmentationfor customer retention extend to customeracquisition . If the fi rm is aware of theprofi tability of new customers beforehand, itcan choose its level of marketing investmentaccording to the expected return from thiscustomer. Yet another strategy consists ofApplying decision trees forvalue-based customer relationsmanagement: Predicting airlinecustomers ’ future valuesReceived (in revised form): 15th January, 2007Giuliano Tirenniis a Post Doc at the Center for Business Metrics at University of St. Gallen, Switzerland.Christian Kaiseris a Post Doc at the Center for Business Metrics at University of St. Gallen, Switzerland.Andreas Herrmannis a Director at the Center for Business Metrics at University of St. Gallen, Switzerland.Keywords marketing , customer equity management , decision trees
Abstract This paper outlines a general methodology for the segmentation of customersaccording to lifetime value and prediction of their future value segments based ondemographic and behavioural features. Not only have few previous scientifi c studies dealtwith this question using nonaggregate single-customer data, but also we address theproblem using decision trees, which has rarely been done before. Applying our methodto customer data from a major European airline produces predictions not only for longterm,middle-term, and short-term customers but also for prospects whose value isdiffi cult to predict because of limited information. Overall, our fi ndings provide marketingmanagers a means to optimise company customer equity and maximise returns onmarketing in the long run.Journal of Database Marketing & Customer Strategy Management (2007) 14, 130 – 142.doi: 10.1057/palgrave.dbm.3250044Applying decision trees for value-based customer relations management© 2007 Palgrave Macmillan Ltd 1741-2439 $30.00 Vol. 14, 2, 130–142 Database Marketing & Customer Strategy Management 131increasing a customer ’ s value for the fi rm byconverting him / her into a member of ahigher profi tability segment. This tactic,again, can be achieved if the main valuedrivers are known to the company. Thus,valuation of customers and knowledge ofcustomer features and behaviour that allowprediction of a customer ’ s value segment areindispensable for optimising marketinginvestments.We obtain satisfying results in predictingcustomer value using decision trees , whichhave some important advantages over othermethods, inclu
本论文由英语论文网提供整理,提供论文代写,英语论文代写,代写论文,代写英语论文,代写留学生论文,代写英文论文,留学生论文代写相关核心关键词搜索。