This study investigates how the mere presence and quality of visual cues affect the popularity of shared accommodations, drawing on cue utilization theory. Using text mining and image recognition methods to process shared accommodation listing data, a multi-method sequential design is applied to examine the combined impacts of pictorial and textual cues related to the listing environment and the host. We also explore the mediating role of social interaction via topic modeling from textual reviews, given its benefits in shared accommodation settings. Results indicate that the mere presence and quality of visual cues each positively influence the listings’ popularity, mediated by customers’ offline social interaction. The effects of pictorial and textual cues differ by presence type and category. This research provides theoretical contributions to the literature on the effectiveness and quality assessment of visual cues in tourism and hospitality. It also offers practical insights for shared accommodation platforms and hosts on optimizing visual cue presentation strategies.