A text analytics approach for online retailing service improvement: Evidence from Twitter

摘要

The purpose of this study is to identify the customers' primary topics of concern regarding online retail brands that are shared among Twitter users. This study collects tweets associated with five leading UK online retailers covering the period from Black Friday to Christmas and New Year's sales. We use a combination of text analytical approaches including topic modelling, sentiment analysis, and network analysis to analyse the tweets. Through the analysis, we identify that delivery, product and customer service are among the most-discussed topics on Twitter. We also highlight the areas that receive the most negative customer sentiments such as delivery and customer service. Interestingly, we also identify emerging topics such as online engagement and in-store experience that are not captured by the existing literature on online retailing. Through a network analysis, we underscore the relationships among those important topics. This study derives insights on how well the leading online retail brands are performing and how their products and services are perceived by their customers. These insights can help businesses understand customers better and enable them to convert the information into meaningful knowledge to improve their business performance. The study offers a novel approach of transforming social media data into useful knowledge about online retailing. The incorporation of three analytical approaches offers insights for researchers to understand the hidden content behind the large collections of unstructured bodies of text, and this information can be used to improve online retailing services and reach out to customers.