Video Design and Product-Link Clicking in Short-Video Commerce: Guide-Oriented Content, Shot-Duration Diversity, and Algorithmic User–Content Fit

Elaborate short-video design may support product-link clicking by providing richer information and more varied presentation; yet, it may also impose greater processing demands during recommendation-driven browsing. This study examines whether two video design features, Guide-oriented content and shot-duration diversity, are associated with users’ likelihood of clicking product links. Drawing on information processing and cognitive load perspectives, we consider semantic and visual temporal processing demands as possible explanations for these associations. Guide-oriented content presents instructional information, whereas shot-duration diversity describes the distribution of shots across duration categories. We further examine algorithmic user–content fit, defined as the platform-estimated correspondence between a video and a user’s historically revealed content preferences, as a boundary condition. Using first-position household-product video exposures from a major Chinese e-commerce platform from 25 July to 23 August 2022, we estimate linear probability models on 66,244 complete-case observations, with date fixed effects and standard errors clustered by creator and user in two dimensions. In the full specification, both design features are negatively associated with product-link clicking. Interestingly, while algorithmic user–content fit also exhibits a direct negative association with clicking, the negative association of Guide-oriented content is weaker at higher fit. Conversely, no robust interaction is observed for shot-duration diversity. The Guide association and its fit interaction become non-significant with user fixed effects, while the diversity association depends on the control set. Creator fixed effects also weaken the evidence for both design features and the Guide–fit interaction. These conditional associations connect video design and platform matching with users’ transition from content viewing to product-page entry.