A Multi-Agent Model of Consumer Behavior Considering Social Networks: Simulations for an Effective Movie Advertising Strategy

Abstract

It is essential for a firm to understand consumer behavior in order to advertise products efficiently on a limited budget. Nowadays, consumer behavior is highly complex because consumers can get a lot of information about products from not only firm’s advertising but also social networking services. The purposes of this study are to construct consumer behavior model considering social networks and to demonstrate an effective weekly advertising budget allocation in order to increase the number of adopters of products. First, we developed a multi-agent model of consumer behavior taking the movie market as an example. In our model, each agent decides whether or not to watch a movie by comparing the weighted sum of “individual preference” and “effects of advertising and word-of-mouth (WOM)” with “individual threshold.” The scale-free network is used to describe social networks. Next, we verified the accuracy of the model by comparing the simulation results with the actual sales figures of 13 movies. Finally, we showed an effective weekly advertising budget allocation corresponding to movie type by simulations. Furthermore, it was demonstrated that the weekly advertising budget allocation gives greater impact on the number of adopters of products as social networks grow.