3. The method
We selected nine viral marketing campaigns to study, details of which are provided in Table 2. From a list of numerous potential candidates, these nine campaigns were chosen on the basis of two main criteria: being global and being successful. A convenience sample was used for individual selection. Success was judged from the initiator's perspective through increased turnover, sales, or brand development, or in terms of how far the message spread. However, the final selection of campaigns could not be a completely random process due to the size of the topic. For example, a Google search of the term viral marketing generated 680,000 hits, far too many for an exploratory search of the relationship between emotion and forwarding behavior. As such, it was necessary to develop an alternative approach to campaign selection. Individual campaigns were identified on the basis of a convenience sample of campaigns that had been seen recently by friends, family, and work colleagues. This approach resulted in a long list, which was then shortened to nine campaigns that were judged to be both global and successful. In summary, two of the nine campaigns selected sought to promote the company (Amazon and e-Tractions), four sought to promote new products (Dr Pepper/Seven Up, Honda, Motorola, and Rock the Vote), two sought to increase awareness (Non-Governmental Organizations [NGOs] and Organization of Women's Freedom), and one sought to save the company (Save BNN).