What to Say and How to Say it: Impacts of Content Type and Linguistic Style on Sales Performance on Live Streaming Platforms

The flourishing live streaming industry has created unparalleled opportunities for e-commerce. While previous studies have explored the effects of content and style of streamers’ language on sale performance, they have largely overlooked the critical interaction between the two. In response, we apply signaling theory to examine the interactive effects of streamers’ verbal content and linguistic styles on sales performance. Drawing on dual-process theory, we analyze two verbal content types—product-relevant and product-irrelevant—alongside two linguistic styles: concrete and psychological intimacy. We also consider live room popularity as a moderating factor. Analyzing 5,160 min of data from a live streaming platform with a fixed-effects model, our findings reveal that both content types are positively associated with sales performance. Product-relevant content conveyed concretely and product-irrelevant content conveyed with psychological intimacy significantly enhance sales. The popularity of live rooms positively moderates the relationship between streamer’s language and sales performance. This research contributes theoretically by extending signaling and dual-process theories to the dynamic context of live-streaming commerce. Practically, it offers actionable insights for streamers and platforms to optimize sales performance based on audience size and content type.