Satisfação, lealdade e retenção: um pré-experimento aplicado à telefonia móvel

• Autor(a) principal: Lopes, José Eduardo Ferreira
• Data de Publicação: 2007
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFU
• Texto Completo: https://repositorio.ufu.br/handle/123456789/12015
• topic: Comportamento do consumidor; Satisfação; Lealdade; Fidelidade; Retenção; Mineração de dados; Redes neurais; Árvore de decisão; Regressão logística; Marketing de relacionamento; Modelagem de equações estruturais; Consumers behavior; Satisfaction; Loyalty; Data mining; Neural nets; Decision tree; Logistic regression; CRM Customer relationship management; CNPQ::CIENCIAS SOCIAIS APLICADAS::ADMINISTRACAO

Satisfaction and loyalty of those faithful consumers to a specific product or brand name are interconnected in a complex way, and, just like CEOs, marketing and statistics academic professionals understand and admit this close relation as well, by publishing some researches which point out the consumers satisfaction as an antecedent of loyalty. However, more often than not, most researches deals with satisfaction and loyalty self-declared by consumers, that is, they analyze this premise under the vision of the customers, who state they are satisfied or loyal. Few are the studies that use consumer s relationship and behavior data, from real organizations databases, to investigate if empirical studies results find any associations between the relational constructs and self-declared behaviors. In this way, this study uses the attitudes and behaviors customers database, of a specific organization, beyond testing satisfaction as an antecedent of loyalty, in the attempt to predict satisfaction, loyalty and customers retention capacity levels of the company. For in such a way, a survey was lead, studying the mobile phones industry. By the appliance of the Structural Equations Model SEM, it was verified that the satisfaction has a positive impact in loyalty, what confirms previous studies. However, from data mining, using Neural Nets, Decision Tree and Logistic Regression techniques, it was not possible to predict satisfaction and loyalty levels since attitudes and behaviors customers database. The analyses results and the academic and management implications are pointed. Also, some suggestions for future studies have been made.