This study aimed to develop a practical method for predicting customer retention likelihood by employing analytical methods different from those used by previous studies. A decision tree (DT) methodology was applied to predict the likelihood of customers not switching to new service providers (NSPs). Because the benefits of using DTs are flexibility and comprehensibility, the DT technique was used to select the items for predicting customer retention likelihood. Empirical data were collected from container shipping customers to demonstrate that the DT technique could be used to develop a customer retention prediction model for the container industry. The results showed that the service attribute of "Container carriers have a very close relationship with shippers" was the covariate with the largest correlation with NSPs. This indicated a close relationship between container carriers and shippers had the greatest influence on a customer who decides not to switch to another NSP. Our results not only suggest a simple decision rule for predicting customer retention likelihood in the container shipping industry, but also provide evidence to support a marketing assertion that customer retention is a central topic in the management and marketing decisions of the industry. Finally, managerial implications are also discussed.
Lin, Le-Hui; Chen, Kee Kuo; and Chiu, Rong-Her
"PREDICTING CUSTOMER RETENTION LIKELIHOOD IN THE CONTAINER SHIPPING INDUSTRY THROUGH THE DECISION TREE APPROACH,"
Journal of Marine Science and Technology: Vol. 25
, Article 3.
Available at: https://jmstt.ntou.edu.tw/journal/vol25/iss1/3