Research On The Impact Of Visual Elements On Sales In The E-commerce Context

The rapid development of the Internet and digital technology has spawned a series of e-commerce platforms,where these platforms play important roles in the national economy.In the e-commerce context,merchants often display a list of product details on web pages.These contents can provide a powerful reference for consumers,information search,alternative evaluation,and decision making.According to the Cone of Experience theory,visual elements are more effective in conveying information compared to text and audio.Therefore,a series of visual elements,such as pictures and virtual reality(VR),are employed in displaying product details online.For the commonly used pictures,most previous literature focuses on examining the economic impact of a certain type of pictures on market performance.However,pictures are classical unstructured data which can demonstrate different contents and features.Few works compare the impact of different types of pictures on sales and explore which visual marketing strategy is the most effective and what is the underlying mechanism of pictures.In addition,as virtual reality is an emerging visual element,few studies conduct large-scale empirical analysis on virtual reality to reveal its influences and mechanisms on e-commerce platforms.The dissertation conducts empirical studies on two kinds of visual elements:"picture" and "virtual reality".The paper treats e-commerce platforms as the research context,collects large-scale data sets,employs advanced econometric techniques to explore the roles of pictures and virtual reality in product sales.Results reveal the underlying theoretical mechanisms and provide managerial suggestions to platform owners,merchants,and consumers.Specifically,the first part focuses on each merchant’s main picture,which is the representative visual element that a merchant displays to consumers.Merchants can display different types of main pictures,such brand name/logo,products,or environment.It is still unclear whether displaying different types of main pictures can generate distinct sales performance.If yes,which aspects can contribute to the sales variation?Thus,the first part compares the impact of different types of pictures on sales,treats a daily-deal platform,Meituan-Dianping,as the research background,employs log-linear fixed effects model and Chamberlin device,and corrects for selection bias to examine the influence of three types of main pictures(brand,product and environment)on sales,revealing the underlying mechanism.The second part focuses on investigating the influence of virtual reality on the selling process of high involvement products on real estate platforms.Virtual reality uses a computer to generate a simulated environment,which enables users to browse information remotely and interact with the environment in real time.Beike,a wellknown online real estate platform in China,introduces its Virtual Reality Tour function in 2018.which is the very first to apply the VR technology in the e-commerce industry.The second part examines the role of virtual reality technology in products’selling process.This study treats the real-estate e-commerce platform as the research background,collects a large dataset of properties with their attributes and sale information and employs survival models,propensity score matching,Heckman’s control function method,and other econometric tools for empirical analyses.Results reveal the acceleration effect of VR on the selling process of properties,demonstrate the roles of VR in consumer’s decision-making process and show the impact of consumer’s experience on VR.Finally,this study investigates the interaction effect of VR and pictures to uncover the relationship between these two visual formats on properties’ selling process.As for the study of pictures,the results indicate that merchants with the brand category of main pictures perform better sales than those with the product category or the environment category of main pictures.The distinction of sales is relevant with two aspects,picture content and picture composition.Picture content denotes the focal objects and elements depicted by the photo,reflecting the information effect.For chain stores with high brand awareness,merchants displaying brand names/logos as picture content can bring more sales than those displaying products or the environment.As for merchants with low brand awareness,the distinction of sales between different types of pictures is insignificant.Besides,picture composition denotes the arrangement of objects and elements to influence the visually compelling,reflecting the aesthetic effect.This paper finds that the aggregate effect of five typical compositional features(i.e.,background ratio,luminance contrast,convexity,symmetry,and enclosure)can explain the sales distinction between different types of pictures.The impact of picture content and composition are heterogeneous to merchants’characteristics.Picture content is more influential for new merchants and less renowned merchants.Picture composition shows significant impacts on both new/old and renowned/less renowned merchants.As for the study of VR,the results show that VR,in general,can accelerate properties’selling process.However,when VR just launches,all consumers do not have prior using experience on it.VR may extend the selling process of properties in a short period.When consumers get familiar with how to use the VR,the acceleration effect of VR will occur and gradually increase.Over time,the acceleration effect becomes stable in the long run.Besides,the presence of the "VR badge" can increase web visits of a property’ s detail page,helping consumers find out the most promising properties within a large set of alternatives during the information search stage.In the evaluation stage,VR-enabled properties have higher attentions.Moreover,VR can increase consumers’conversion from web visits to attention,indicating the effectiveness of information provided by VR may help consumers discern property quality.Finally,our finding demonstrates that VR and pictures present a substitution effect.For a property supporting both VR and pictures,the seller may speed up his property’s selling process by reducing the number of pictures to avoid the potential information overload.The dissertation extends the literature of the impact of pictures on sales in the ecommerce environment,explains the sales variation with different categories of pictures from the viewpoint of picture content and composition and shows the heterogeneous effect of content and composition on sales.Besides,this dissertation uncovers the role of VR in consumers’ decision-making process,demonstrates the influence of consumers’ prior experience on VR and shows the relationship between pictures and VR.The findings can enrich theories and literature related to e-commerce,visual marketing,high involvement products,virtual reality,consumer decision-making process and provide managerial implications for e-commerce platform owners and merchants to better employ pictures and VR for online presentations.