The influence of emotional tone on user engagement with media brands on X

Abstract

Understanding what drives engagement on social media is a key challenge for media brands seeking to strengthen relationships with digital audiences. This study investigates the influence of emotional tone on digital customer engagement on X by analysing interactions between users and major Colombian media brands. Using a dataset of more than 100,000 Spanish-language tweets collected over a one-year period, the research examines how specific emotions — calmness, sadness, anger, and fear — affect engagement behaviours measured through likes, retweets and replies. Advanced natural language processing and machine learning techniques were employed, combining the BETO and ROBERTA models to classify sentiment and emotional tone, while thematic analysis was used to identify content categories including politics, entertainment and sports. The findings indicate that emotional tone is strongly associated with engagement patterns. High-arousal emotions, particularly anger and fear, generate the greatest levels of active engagement through retweets and replies, whereas calmness and sadness are more closely linked to passive engagement through likes. Political content was found to attract the highest levels of interaction, while entertainment content produced more moderate responses. The study highlights the importance of considering both emotional tone and emotional intensity when developing social media strategies. These findings provide practical guidance for media brands seeking to optimise audience engagement through emotionally resonant content while contributing to the growing literature on emotion-driven digital engagement. This article is also included in The Business & Management Collection which can be accessed at https://hstalks.com/business/.