The Impact of AI-Recommended Content Affordances on Post-Purchase Intention in Stockout Substitution Scenarios

Stockouts significantly threaten consumer loyalty and cause substantial economic losses. In response, online platforms are widely deploying AI recommender systems to provide substitutes. However, whether such AI-driven substitution strategies can effectively mitigate the negative consequences of stockouts remains underexplored. Grounded in technology affordance and perceived value theories, this study develops a conceptual framework to investigate how content affordances of AI-recommended substitutes-specifically perceived fit, personalization, and serendipity-influence post-purchase intentions through functional and emotional value perceptions. Analysis of survey data from 479 respondents reveals that these affordances enhance perceived value, which in turn strengthens post-purchase intentions. Moreover, the findings demonstrate distinct effects of each affordance dimension on perceived functional value versus emotional value. In terms of the moderating effects, privacy concerns positively moderate the relationship between perceived functional value and post-purchase intention. Necessary Condition Analysis (NCA) further identifies critical prerequisites for achieving high perceived value and post-purchase intentions. This study extends the application of AI recommender systems to service recovery contexts and offers a wealth of novel insights for designing effective substitution strategies.