Measuring and Managing Consumer Sentiment in an Online Community Environment

Abstract

As social media and virtual communities increase in popularity, the spread of word of mouth becomes easier, challenging firms to measure and manage the success of marketing initiatives in online community environments. This research examines how consumers react to firms' active participation in consumer-to-consumer conversations in an online community setting. The authors develop a tailored community-matched measure of consumer reaction (consumer sentiment) and analyze more than 115,000 consumer posts from ten online forums with active firm participation. The results indicate that consumers show diminishing returns to active firm engagement, which, at very high levels, can undermine consumer sentiment. Further subgroup analyses by conversation type indicate that these relationships hold for conversations that address consumers' functional needs but do not hold for conversations that address social needs. Finally, the results show diminishing returns to firm engagement for consumers primarily interested in product-related support but show no relationship for consumers primarily interested in inspiration and entertainment. These findings provide insights for marketing performance measurement and resource allocation in online communities.

Gender bias in sentiment analysis

“…It takes advantage of the public availability of opinionated texts and fast, accurate software for detecting opinions. It can also be used to assess the impact of business interventions in the social web (Homburg, Ehm, & Artz, 2015).…”

Making sense of consumers’ tweets

“…Therefore, eWOM and user-generated contents (UGC) emerge as key drivers for consumers' buying decisions (Gensler et al, 2015;Lee et al, 2012). For these reasons, a large body of literature investigates the effect of eWOM on sales (King et al, 2014), other authors tried to identify the effect of positive or negative online comments/posts/reviews (Hornik et al, 2015;Li et al, 2013;Pang and Qiu, 2016;Yang et al, 2015), the best strategy to influence consumers' positive eWOM (Erkan and Evans, 2016;Reamer et al, 2016), consumers sentiments towards a certain brand (Mostafa, 2013); and consumers' reactions towards the firm's participation in their online conversations (Homburg et al, 2015), while other studies have further considered using online consumers' rates of tourism attractions as predictors of their future behaviour . Thus, the online contents generated by consumers in terms of reviews, ranking and ratings largely convey more information than reputation, in other words they reflect the product quality and popularity of a certain product, brand or service (Li et al, 2013;Liang et al, 2015;Tsekouras, 2017).…”

Social Media Posts on Samsung Galaxy Note 7 Explosion: A Comparative Analysis of Crisis Framing and Sentiments in Three Nations

“…Sentiment analysis has been used widely in social media research because it can measure feelings toward communication in an interface. Sentiment on social media is defined as the attitudinal valence of users' mentions (Homburg, Ehm, & Artz, 2015). In other words, social media posts about a crisis in terms of sentiment can be viewed as negative, positive, or neutral (Choi & Lin, 2009).…”