Article Preview
TopIntroduction
With the explosive growth of online sales in today’s global economy, the Internet has offered a much broader market space for the sellers as well as for the buyers. For example, TaoBao, the top Chinese online mall, announced that its sales reached 35 billion RMB in a single day, November 11, the “double sticks” day in China in 2013.
However, unlike the traditional markets where sales are conducted face-to-face between the sellers and the buyers, the online sales tend to be rather 'anonymous' with little personal interaction (Cabral & Hortacsu, 2010; Ye, Li, Kiang, & Wu, 2009). This may result in intensified information asymmetry leading to potential risks for the buyers (Valentin Andrei, 2011). As the key stakeholders in online sales, the buyers have the ultimate rights for decisions concerning whether to buy, what to buy and where to buy. To mitigate the risks and uncertainties confronted by the buyers within an online sales context, trust in the sellers is a necessity for the buyers to accept any risk associated with the online sales (Gefen & Straub, 2004; McCole, Ramsey, & Williams, 2010). As a result, it has been well recognized that the lack of trust between the buyers and the sellers explains numerous failures of online transactions (Grabner-Kräuter & Kaluscha, 2003). Establishing an effective trust system is crucial for the development of e-commerce to reach its full potential.
To this end, the seller reputation mechanism has been built for collecting information from the seller's transaction history (Resnick, Zeckhauser, Swanson, & Lockwood, 2006). This seller reputation information is available to potential buyers, thus has a considerable effect on the buying decision. For the online sales market with a high degree of uncertainties for the buyers, seller reputation is an important asset for the sellers as the basis of the buyer’s choice for the sellers. Therefore, the seller reputation mechanism plays a key role in the sellers’ effort to build and maintain a good reputation as well as in the online sites’ endeavor for the online sales prosperity (Bar-Isaac & Tadelis, 2008).
The dominant online sales site TaoBao, for example, has set up a buyer feedback system for all the sellers. After the completion of each transaction, the buyer is asked to grade the seller reputation in terms of the truth of the goods description, the service quality, the goods delivery speed, and so on. In addition, the buyers can give a subjective evaluation about the transaction using comments in text. A potential buyer is able to access the seller reputation information before the buying decision can be made. Thus, the seller reputation mechanism plays a crucial role in the context of online market where each buyer may be faced with mushrooming number of possible sellers (Resnick et al., 2006). However, it has been noted that the reliablility of such seller reputation mechanism may be compromised due to potential fraudulent behaviors of the buyers as well as few feedbacks from the consumers (Dellarocas, 2000). This poses an interesting and very important question: how well do these seller reputation systems work?
This paper attempts to investigate the effectiveness of seller reputation mechanism by an empirical study using data collected from TaoBao.com, which is the largest online market in China. After an extensive analysis of the seller reputation metrics for the TaoBao seller reputation system, the small appliances are selected as the input for the empirical study, which are normally standardized products with a fairly good product homogeneity among different sellers of the market. In order to investigate the effectiveness of various seller reputation metrics, the univariate analysis is carried out for revealing the impact of various seller reputation metrics on the buyers’ decision. In addition, the attribute selection technique is applied for finding out the most significant factors contributing to the sales performance. The results of this study will provide empirical evidence for improving the seller reputation systems.