Abstract
With the rapid development of the internet, social media platforms enable tourists to extensively research, compare, and evaluate potential destinations. Concurrently, these platforms have provided new channels for destinations to present themselves and engage with potential visitors. Therefore, the closer destinations’ selfpresentation is to the actual experience of the tourists, the higher the likelihood that tourists’ expectations are met. This study analyzes image posts from tourists as well as DMOs of popular cities around Lake Constance on Instagram. An image content recognition technique and a clustering algorithm is applied to obtain insights into the preferences of tourist segments within individual destinations. These insights are then compared to how the destinations present themselves on Instagram. The results demonstrate that the data reveals distinct segments within the tourist groups of the studied destinations. Furthermore, the comparison between the external perception and self-presentation shows a high perception congruence of similarity for two destinations but also significant discrepancies for one destination examined.