PERSUASION IN REWARD-BASED CROWDFUNDING: THE ROLE OF VISUAL IMAGE

As a new form of fundraising for business startups, crowdfunding (CF) benefits the entrepreneurs by (1) overcoming geographical constraints (Agrawal et al., 2011); (2) adding publicity (e.g., Belleflamme et al., 2013); (3) enabling them to collect “wisdom of crowds” (Collins & Pierrakis, 2012). To attract support from potential backers, entrepreneurs need to strategically share sufficient and effective information about their business and themselves. Given the picture superiority effect (Childers & Houston, 1984), the research about visual information and text-picture integration is missing. Take restaurant reward-based CF as an example, this research is designed to fill the research voids by answering to the following research questions: (1) What are the major themes of visual information used by the entrepreneurs? (2) How do potential backers respond to the (combinations of) visual information used on the projects? (3) How do entrepreneurs articulate identities on their profile? (4) How do potential backers respond to the incongruence between the entrepreneurs’ textual and visual self-presentation? Drawing upon the signaling theory, Study 1 identifies visual signals captured from the main images of all existing restaurant reward-based CF projects on Kickstarter using Google Vision API and conducts a series of hierarchical clustering analyses using KH coder. The effectiveness of the substantive visual signals (i.e., prototype/team/asset) and their combinations with the rhetorical signal (i.e., logo) is further investigated using an experiment. As hypothesized, the visual signal of the entrepreneurial team is found to evoke superior perceived expertise than other signals; the visual signal of the asset is found to induce more liking than other signals when the logo is present. By clustering analyses on the Kickstarter profiles of all existing restaurant entrepreneurial teams, Study 2 constructs two personas of the teams – “neighbor” vs. “businessman”. Combining two types of personal photographs differentiated by their outfits (i.e., casual vs. professional), this study further investigates the effectiveness of text-picture integration in the entrepreneurial self-presentation using an experiment. As hypothesized, the incongruence between the textual self-presentation (i.e., the persona) and the visual self-presentation (i.e., the personal photographs) has significantly positive effects on the message effectiveness, potential backers’ attitude, and their intention to support.