Recognition of Profitable Customer by Neural Networks

摘要

Recently, the development of marketing strategy is changing from customer oriented to relationship oriented. Customer centricity is the pivotal strategic issue in business relationship development. Meanwhile, customer centricity is all about growing enduring relationships with profitable customers, and retaining important customers is crucial to business success. Therefore, how to retain profitable customers is the critical issue of relationship marketing. Any business that develops strong relationships with their customers over the long-term, possess a unique advantage over their competitors. Segmentation is one of the most useful methods to discover profitable quotient. However, segmentation of profitable customers is specific to the context of different business, and it is difficult to be recognized in models. In the research, a method is developed to recognize profitable customers by back-propagation neural networks. First, exploratory research including field notes is applied to the association of dentistry clinic to identify all parameters of profitable customers and influencing factors of resulting profitable customers. Second, set of parameters of profitable customers is trained by one class of neural networks, the multiplayer feed-forward neural networks with sigmoid function training by back-propagation training algorithm. Then, profitable customers are recognized from the influencing factors, where these factors are the information of each customer. This method would properly recognize profitable customers, which could be used on marketing planning. After identifying profitable customers, the finial strategy will be evaluated by introducing into the operation of the association of dentistry clinic in southern Taiwan. Therefore, competitive marketing strategies, which can fit better in the competitive business environment could be made.