unique potential to reach large numbers of potential customers in a short period of time at a lower cost. While some VM campaigns take off, others fail. For every high-profile example of a successful VM campaign, there are many more VM attempts that fail (Watts & Peretti, 2007).
Implementing VM campaigns, marketers often face substantial challenges. In the case of Subaru’s ‘First Car Story’ campaign, for instance, if the campaign does not take off quickly, what is the reason and what should Subaru’s marketers do? Should they increase the size of seeding (the initial set of targeted consumers chosen by the initiator of VM campaign)? Should they better match the content of the message to the target market? Should they spend more on the programming of the video application to make it easier and more attractive? Or is the problem the mechanism for including the resultant video into social media? Marketing efforts addressing each of these issues will likely pay off to some extent, but most marketers have limited resources. Thus, it is important for marketers to first find out where in the VM process things are failing. A remaining and critical question is ‘Which of these issues should receive more attention than others to pass the threshold for a propagation of a marketing message throughout a target market?’
Despite increasing popularity of VM, there is little research that provides the foundation that marketers should bring to an analysis of a failing VM campaign. Most of the recent research on VM focuses on one component of VM campaigns (e.g., social network structures, seeding, etc.). Although these studies are helpful in understanding and improving a certain component of a VM campaign, as a result of focusing on one or two aspects of VM, they provide little insights into what the various components of a VM campaign are and how they must come together. The purpose of the present research is to help marketers take a holistic approach to a VM campaign and understand what can drive a propagation of a viral message. We provide a framework with which marketers can estimate their VM campaign and help them effectively decide where to direct their marketing efforts. To this aim, we develop a VM model based on epidemiology, the study of spread of disease in populations, and analyze relative impact of VM components on VM success or failure. We achieve this objective by conducting sensitivity analyses through computer simulations with various parameters under different scenarios and conditions.