Investigating Consumer Attitudes Toward Recessive Advertising in Short-Form Videos

Abstract

In we-media era, recessive advertising frequently appears in the user-generated contents, especially in the short-form videos. Yet we know little about whether consumers are able to recognize recessive ads and their responses to recessive ads. In this study, we conduct data mining to explore how consumers recognize recessive ads and their attitudes toward these ads. We use Python to crawl 178,000 structured bullet comments from three short videos in a representative Chinese short-form video app Bilibili.com. We first extract the common features of recessive ads (suddenness, persistence, familiarity) by visualizing the bullet comments. Thereafter, E-DIAF model is developed to explore how the common features affect consumers’ identification of recessive ads. Additionally, we establish a Multidimensional Emotion Computing Model (MDE-CM) to conduct sentiment analysis, which uncovers consumers’ significant emotional shifts, especially the negative emotions such as “badness” and “fright” during ad segments. Findings indicate that seamless integration of ads and video content could minimize consumers’ negative emotional responses toward ads, meanwhile enhance brand awareness and affinity. Theoretical and practical contributions are discussed.