Too Big to Sell? A Computational Analysis of Network and Content Characteristics among Mega and Micro Beauty and Fashion Social Media Influencers

Abstract

Fashion and beauty brands leverage social media influencers to shape purchasing decisions, improve cost effectiveness, and reach wider audiences. New conventional wisdom has brands moving away from megainfluencers toward microinfluencers due to greater perceived relatability and trustworthiness. This study employs a novel computational approach integrating network analysis and computational text analysis to understand differences in content and its diffusion through mega- and microinfluencer Twitter networks. Findings debunk conventional wisdom that microinfluencers can best fill unique roles by forging intimate, emotion-laden interpersonal connections. While microinfluencers are more central to two-way dialogue within their networks, megainfluencers garner more affect directed toward them, indicating greater trust. Practical implications for the continued value of megainfluencers and the identification and development of promising microinfluencers are discussed.

Keywords:

• Social media influencers
• influencer advertising
• computational research methods
• beauty and fashion industries
• Twitter

Additional information

Notes on contributors

Rebecca K. Britt

Rebecca K. Britt (PhD, Purdue University) is an associate professor, Department of Journalism and Creative Media, University of Alabama.

Jameson L. Hayes

Jameson L. Hayes (PhD, University of Georgia) is an assistant professor and director of the Public Opinion Lab, Department of Advertising and Public Relations, University of Alabama.

Brian C. Britt

Brian C. Britt (PhD, Purdue University) is an assistant professor and associate director of the Public Opinion Lab, Department of Advertising and Public Relations, University of Alabama.

Haseon Park

Haseon Park (MA, University of North Dakota) is a doctoral student, Department of Advertising and Public Relations, University of Alabama.