Understanding engagement with Instagram posts about tourism destinations

This study analyses the social media engagement (SME) received by Instagram posts from a tourism destination based on communication and mental imagery theories. This research considers both the type of sender (tourists vs. residents) and the content of the message (pictorial stimuli: places of interest vs. hospitality services; centricity stimuli: people vs. without people). Web scraping technique is used for data collection. Content analysis is then applied on 27,088 Instagram posts using artificial intelligence techniques (machine learning and deep learning); and a univariate generalized linear model is conducted to analyze differences in SME. Results show that pictorial stimulus determines SME, being higher for images focused on places of interest, and photographs with people get higher SME too. The type of sender also influences SME and exerts a moderating role reinforcing the effect of centricity on SME for tourists. These results provide interesting insights for destination marketing managers and Instagram users. • 139,273 UG-photos about a destination in Instagram are analyzed using AI. • Machine and deep learning are used to analyze the photo’s text and visual content. • We differentiate between point of interest and hospitality service photos. • Point of interest photos double the engagement of hospitality service ones. • Photos including people and posted by tourists get more engagement.