A Fuzzy Multiattribute Approach to Help Measure Quality of Online Classifieds Systems
Ahmet Beskese and Tolga Sen
In e-commerce systems, the products are stationary and centric by nature. Hence, many of the traditional quality metrics should be interpreted in a different manner in practice. One of the most important reasons for such a different interpretation is that the focus must be on the manageability and quality of data for the systems that are interactive and dense by data. Online classifieds systems constitute a special research area in e-commerce systems because of their non-payment nature. However, there is only a very limited amount of research dedicated to defining their quality if any at all. This study aims at proposing a model to help measure quality in online classifieds systems. The model is also expected to guide managers in their quality improvement activities by defining priorities. Since the experts preferred natural language expressions rather than sharp numerical values in assessing quality parameters of an online classifieds system, and there are some incomplete evaluations, a slightly modified AHP method adopted from Zeng et al. is used for the prioritization problem at hand. To illustrate how the model works, a real-case application is provided from Turkey’s leading e-real estate organization. In this example; penetration score which basically includes the average results of the routine security penetration tests is found to be the most important quality factor. This factor is followed by usability and response satisfaction (customer satisfaction rate for ticket responses). The results achieved constitute a very good starting point for the managers of online classifieds systems who try to allocate their scarce resources in the most efficient composition.
Keywords: Quality measurement, online classifieds, fuzzy evaluations, AHP