Read This, Please? The Role of Politeness in Customer Service Engagement on Social Media

Abstract

People are increasingly turning to social media for help. According to a recent report by Twitter, over 5.5M customer service-related tweets are generated per month. In this work, we aim to explore firms' strategy when engaging customers regarding their concerns and complaints on Twitter. Specifically, we focus on how politeness, a linguistic factor indicating how a customer is questioning or complaining rather than the content of a query, affects firms' customer service engagement strategy. We develop a novel text mining methodology to mine politeness from tweets. Using this approach, our estimation results show several interesting results, including that firms are more likely to respond to more polite customers, and that this effect is augmented for customers with high social status. However, firms are more likely to engage impolite customers with a high social status in a private channel such as through direct messaging.