Essays on Online and Multi-Channel Marketing

Firms increasingly adopt online and multi-channel marketing strategies to reach and persuade consumers. Therefore, designing an effective marketing mix is critical to their success. The aim of my dissertation is to understand the strategy behind firms’ channel choices and assess marketing effectiveness. It consists of three large-scale empirical studies examining several important aspects of online and multi-channel marketing. My first essay focuses on the business-to-business (B2B) interactions involving online platforms, which serve as new channels for traditional merchants to reach consumers and grow business. Using data from the daily deal market, we specify a structural model that examines consumer choices on the demand side and firm strategies on the supply side. In particular, we incorporate merchant heterogeneity and allow prices to be jointly determined by merchants and platforms through negotiation; both of these match the real-world complexity but are challenging to be modeled theoretically. Our results show how platform size, commission rate, and the allocation of price-bargaining power jointly determine the price setting and the platform differentiation among merchants. Essay two studies to what extent marketers’ actions can affect the reach of video advertising campaigns through influencing the amount of user-generated content. To do so, we compile a unique and comprehensive data set on ad campaigns conducted on video sharing sites such as YouTube. We find that several instruments under the control of advertisers can influence how much the reach of a campaign benefits from user-generated content. Our results underscore that, with the right strategy, advertisers can substantially increase the number of impressions that their online video campaigns yield. Essay three assesses the effect of advertising and personal selling in the U.S. presidential elections, where advertising involves both candidate campaign ads and those sponsored by outside political interest groups and personal selling takes the form of field operations. We set up a structural model that treats campaign allocation as endogenous and also allows the campaign effect to vary across individuals. Among the many findings, we show that field operations are more effective for partisan voters whereas candidate campaign ads are effective for non-partisans. Interestingly, ads from outside political groups are more effective for partisans than for non-partisans. Our counterfactual results indicate that field operations play a critical role in the 2008 and 2012 elections while the importance of ads is only substantial in a close competition like the 2004 election.