As competition intensifies, retaining customers becomes one of the most serious challenges facing customer service providers. Customer attrition prediction models hold great promise as powerful tools for enhancing customer retention. Several statistical methods have been applied to develop models predicting customer attrition. Yet little research is done on the relative performance of models developed by different methods. The lack of knowledge about the performance of various prediction models is more pronounced due to the nonlinear nature of the combined causes of attrition (such as switching to another provider or canceling a service). The development of data mining techniques has made the comparison of prediction power of different models more efficient and easier. In this article we demonstrate how to use data mining techniques and software to fit and compare different customer attrition prediction models, using data from a major telecom service provider.
This paper identifies information sources and practices of environmental scanning preferred by managers of globally oriented small and medium-sized enterprises (GOSMEs). Data were collected using a Delphi technique and were analysed by NUD*IST software and the Homogeneity Analysis technique. Major findings indicate that although managers of GOSMEs generally prefer external and personal sources in their environment scanning process, contingent conditions related to the industry, the organization and the owner-manager guide the choice of appropriate information source and the need to scan systematically each sector of the environment. Statistical relationships were identified, and these relationships allowed the formulation of general propositions that could be helpful for practice and research in GOSMEs. The paper concludes that the manager's need to scan systematically a specific sector of the environment and the information source the firm might use are dependent on the level of uncertainty aroused by this sector, the amount of pertinent information the source has, and its accessibility by the firm.
Considering the subsidiaries of multinational companies, a study of the conditions affecting the strategic choice is conducted. Beginning with an analysis of the taxonomies of subsidiaries, two strategic models were selected to test three hypotheses: one, regarding the variables that influence the strategic choice; a second, about the relationship between the two classification models and a third, about the influence of the type of subsidiary on performance. This study involved five European countries, namely the United Kingdom, France, Germany, Sweden and Portugal. It concludes that nation based variables have important influence in the determination of subsidiary strategic roles. This influence is present especially in terms of national culture with variables like individualism and uncertainty avoidance, while analyzing relationships, and age and technological capacity, while analyzing functions. We also found that both models used are independent, suggesting that they are complementary models and not substitutes.