Applying User Interface Analytics to Identify Online Shop Performance Factors

Abstract

The massive use of information systems and digital applications drives the growth of e-commerce, including online shops in marketplaces. However, some of the online shops are not successful. To improve their performance, the success factors of the online shops should be recognized. This study develops a model of online shop success factors. Unlike the other researches that use customer preference data from surveys, this study uses user interface analytics to develop the model. A marketplace operated in Indonesia was selected as the case study. The study begins with a scraping process of the data available at online shops' user interfaces in the marketplace. After data cleaning, outliers handling, and data clustering by product category, a series of multiple regression analyses are performed to get the model estimates. Eight variables are defined to develop the model, i.e., product price, percentage of responded chat, shop joining time in the marketplace, number of product types, number of raters, shop rating, shop reputation, and number of followers. The results of the multiple regression process show that the model estimate is specific for every product category. The final model can be used as a reference by the online shop sellers to develop their strategy to improve their shop performance. Moreover, the results also prove that user interface analytics is effective in estimating the performance factors of online shops in a marketplace.