Modeling Customer's Satisfaction Behavior through Uninorms

Abstract

During the last three decades, the focus of customer satisfaction research has shifted from what it was about the product or service that customers found satisfying to how and why customers became satisfied. This resulted into several models that try to explain the customer's satisfaction behaviour, among which the expectancy-disconfirmation paradigm is one of the most prominent models. This model identifies three elements which have an influence on the customer's satisfaction level: i.e performance, expectation and disconfirmation. The customer evaluates the perceived product/service performance against a personal norm, which is mostly referred to as his product/service expectation. The expectation level is considered to have a direct proportional effect on the customer's satisfaction. The discrepancy between perceived performance and expected performance is called the expectancy disconfirmation and also has a direct proportional influence on the satisfaction level. Some authors believe that performance also has a direct influence on satisfaction, while others assue that performance, only has an indirect influence through the disconfirmation process. Furthermore, it is most likely that customers perform the entire expectancy-disconfirmation process at a product/service attribute level first, followed by an aggregation to the overall satisfaction level. This aggregation, performed unconsciously by the customer, exhibits both reinforcement and compensation behaviour. Measuring a customer's expectation directly can be a troublesome process. Several levels of expectation exist and it's not always clear which type of expectation a customer uses as his norm. Furthermore, post-consumption measurement of expectation can be biased because product/service experience influences a customer's pre-purchase expectation. However, uniforms can be used to model the expectancy-disconfirmation paradigm, which allows us to derive a proxy for the norm mathematically. It can be shown both theoretically and empirically that a close match exists between the uninorm's properties and the expectancy disconfirmation process. The uninorm aggregator can be written mathematically as the combination of a generator function and its inverse. Every displaced generator function implies a new uninorm with a different neutral value. This allows us to build the uninorm mathematicall for each customer individually, based on a single predefined generator function. The uninorm's neutral element then acts as the customer-specific norm in the modelled expectancy-disconfirmation process. The usability of the uninorm-derived customer's norm seems very promising. It offers new insights into a company's performance and helps to predict loyalty among a company's customer base, without having to deal with the problems of a directly measured expectation.