Abstract: In the current highly fragmented (social) media landscape, brands must increasingly rely on their own consumers to reach out to new potential consumers. Taking a social networks approach, the current study examines patterns of consumer engagement with brands on social media, identifying the major clusters as well as key users and messages that are responsible for cross- clusters engagement and information flow. After identifying these clusters, we employ large-scale automated content and image analysis to (a) identify the social mediators, and (b) the content characteristics (i.e., media richness, concepts expressed & brand emphasis in images) and sentiment of messages that cut through clusters. Our initial findings suggest that content that gets mediated across clusters is generally less marketer-driven and less polished than traditional advertising, possibly grassroots-style. These results highlight the novelty, feasibility and value of the proposed approach to explaining brand-related content flow across clusters.
