Working Paper
How do opinion leadership and social contagion affect the adoption of a new product? Marketers have been experimenting with viral network marketing based on assumptions that social contagion is at work, that certain customers’ adoptions and opinions have a disproportionate influence, and that firms are in fact able to identify and target opinion leaders. Recent research, however, has raised conflicting opinions about how important opinion leaders are in accelerating the acceptance of new products, whether marketers should identify opinion leaders based on self-reports or on their centrality in social networks, and whether heavy users of a product are more influential than light users.
This study’s research setting was the diffusion of a new drug for the treatment of a chronic and potentially lethal medical condition among physicians in three cities. There are several findings. Contagion operated over network ties and was affected by peers’ usage volume rather than by whether peers had adopted the new drug. Self-reported opinion leaders were less responsive to their peers’ behavior, while sociometric leaders (i.e, those identified by their centrality in social networks) were not differentially responsive. Heavy use at the category level before the new product’s launch was associated with early adoption. Both self-reported leaders and sociometric leaders tended to adopt early, with the tendency being more pronounced for sociometric leaders. Self-reported leadership and sociometric leadership were only moderately positively correlated.
The evidence from this study supports the use of network-leveraging campaigns that focus on central opinion leaders. The study also suggests that the industry practice of overweighting marketing efforts at launch on heavy users is sound. The authors’ findings about sociometric versus self-reported opinion leadership and about contagion being moderated by usage volume suggest both ways to increase theoretical understanding of social contagion dynamics, and ways through which marketers might increase the effectiveness of network marketing.
About the authors
Raghuram Iyengar
is Assistant Professor
of Marketing and
Christophe Van den
Bulte is Associate
Professor of Marketing,
both at the Wharton
School, University of
Pennsylvania.
Thomas W. Valente is
Professor of Preventive
Medicine at the Keck
School of Medicine,
University of Southern
California.
This working paper was featured in Knowledge@Wharton.
http://knowledge.wharton.upenn.edu/article.cfm?articleid=2170
Comments from members
Please login to view and/or submit comments