One of the main objectives of data mining is to help companies determine to which potential customers to market and how many resources to allocate to these potential customers. Most previous works on competitive influence in social networks focus on the first issue. In this work, our focus is on the second issue, i.e., we are interested on the competitive influence of marketing campaigns who need to simultaneously decide how many resources to allocate to their potential customers to advertise their products. Using results from game theory, we are able to completely characterize the optimal strategic resource allocation for the voter model of social networks and prove that the price of competition of this game is unbounded. This work is a step towards providing a solid foundation for marketing advertising in more general scenarios.