Search

Thesis (Ph.D.)--Massachusetts Institute of Technology, Sloan School of Management, 2002., Includes bibliographical references., This thesis contains three separate essays that deal with word of mouth. In the first essay, "Promotional Chat on the Internet," we analyze the firms' incentives to anonymously supply positive reviews of products in chat rooms and other recommendation sites. This, in turn, lowers the credibility of word of mouth transmitted online. We develop a game theoretic model where an incumbent and an entrant that are differentiated in quality compete for the same online market segment. The consumers are uncertain about the entrant's quality, whereas the firms know the value of their products. The consumers hear messages online that make them aware of the existence of the entrant as well as help them decide which product is superior. We find a unique equilibrium where online word of mouth is informative despite the promotional chat activity by competing firms. In this equilibrium, we find that firms spend more resources chatting up inferior products. We also find that promotional chat may be actually more beneficial to consumers than a system with no promotional chat. In the second essay, "Using Online Conversations to Measure Word of Mouth Communication," we test a long-held belief that word of mouth recommendations have a tremendous influence on the sales of new products. So far, there has been little empirical evidence to support this belief since, before the advent of the Internet, word of mouth recommendations were exchanged in private conversations that left no documentary evidence. The Internet provides a window into some of these private conversations and thus a means of measuring word of mouth activity., (cont.) We pose the following pragmatic question: can we use these data to measure word-of-mouth and predict future product sales? We develop and test a model that predicts which metrics of online discussion activity should be correlated with long-run performance. Our empirical findings demonstrate that certain measures of online word-of-mouth are predictive of sales though their predictive power varies significantly over the show's lifetime. Finally, in the third essay, "The Influence of Social Networks on the Effectiveness of Promotional Strategies," we examine the role of social network in word of mouth. The defining characteristic of "buzz" strategies is that the sellers approach the consumers directly, either in online chat rooms or in physical locations such as cafes or nightclubs. The goals of such strategies are twofold: to turn the approached consumer into a buyer and a missionary. The basic characteristic of buzz is that its diffusion is dependent on one's neighbors in the network in contrast to advertising, which allows the firm to communicate with consumers independently of their neighbors. We examine the network's moderating effect on the payoff from firm's investment to promote buzz. In addition, we compare the effectiveness of buzz promotion versus mass advertising as stand-alone marketing instruments and also analyze an advertising campaign that consists of both instruments., by Dina Mayzlin., Ph.D.