2010- Working Papers: Marketing
A comparative and cumulative meta-analysis of advertising appeals, 41 pp.
J. Hornik and G. Miniero
(Working Paper No. 7/2010)
Marketing managers and scholars have focused substantial attention on the role of advertising message appeal in the persuasion process. We conducted a comprehensive quantitative and qualitative literature review of more than a thousand reports, not previously systematically reviewed, to study whether and by what size advertisements with appeals are more persuasive and better liked than ads with no structured appeals. We compared the most common appeals (fear, humor, sex, comparative, gain/loss frame, two or one sided, and metaphor) to determine their relative effectiveness. We also performed separate analyses for each response variable to assess the contribution of several moderating effects on the results. A test of heterogeneity indicated the presence of moderators on observed relationships. Moreover, we employed a cumulative meta-analysis (CMA), at every point during the history of a research domain to test for stability of results. To supplement the quantitative analyses and to analyze qualitative papers and case studies, we employed a qualitative comparative analysis (QCA) on an additional 306 qualitative reports. The meta-analysis included 703 articles containing 1631 independent samples and 2547 effect sizes representing 142,443 participants. Results show that the overall appeal effect between conditions is of small to moderate size and that profound differences exist between appeals while some moderating variables have significant effects on effect sizes in appeal studies. All methods provided the opportunity to obtain results of theoretical and practical interest. We discuss explanations for these results, suggest managerial applications, and identify directions for future research.
The social aspects of customer churn, 47 pp.
I. Nitzan and B. Libai
(Working Paper No. 10/2010)
Customer defection is of much concern to firms and has been the focus of numerous studies, yet in contrast to customer acquisition (new product adoption) its social aspects are largely unexplored. In this study we aim to assess the role of a customer’s social network in his or her defection decision. To do so we use a database of more than one million personal customers of a cellular company. Using data on the communication among customers, we were able to create a large-scale social system composed of the individual social networks of those customers. Our data set includes the entire relevant customer base, which helps us to avoid the non-trivial problems associated with sampling from large-scale networks. We estimate a hazard model that combines social network information with data on customer retention to assess the extent to which customers’ defection from a cellular provider might be affected by their exposure to defections of their social neighbors. Our findings indicate that a customer’s social network affects the defection decision significantly; exposure to a defecting neighbor is associated with an increase of up to 150% in one’s hazard of defection, and 80% when controlling for a host of social, personal and purchase related variable. Furthermore, we show how strength of ties and the degree of similarity (homophily) to defecting neighbors have a significant effect on this phenomenon, and observe that the social defection effect decays exponentially with the passage of time following a neighbor’s defection event. We also demonstrate how customer loyalty moderates the social effect on retention. These results carry important implications for our understanding of the drivers of customer defection, and they should be taken into account by firms that aim to predict and affect the retention of their customers.
When Kerry met Sally: Politics and perceptions in the demand for movies, 43 pp.
J. M.T. Roos and R. Shachar
(Working Paper No. 14/2010)
On election days many of us see a colorful map of the U.S. where each tiny county has a color on the continuum between red and blue. So far we have not used such data to improve the effectiveness of marketing models. In this study, we show that we should. We demonstrate the usefulness of political data via an interesting application—the demand for movies. Using box office data from 25 counties in the U.S. Midwest (21 quarters between 2000 and 2005) we show that by including political data one can improve out-of-sample predictions significantly. Specifically, we estimate the improvement in forecasts due to the addition of political data to be around $43 million per year for the entire U.S. theatrical market. Furthermore, when it comes to movies we depart from previous work in another way. While previous studies have relied on pre-determined movie genres, we estimate perceived movie attributes in a latent space and formulate viewers’ tastes as ideal points. Using perceived attributes improves the out-of-sample predictions even further (by around $93 million per year). Furthermore, the latent dimensions that we identify are not only effective in improving predictions, they are also quite insightful about the nature of movies.
The effect of usage uncertainty on mobile phone package size and subscribers’ choices, 28 pp.
R. Iyengar, O. Koenigsberg and E. Muller
(Working Paper No. 16/2010)
Research No. 04800100
Wireless services typically charge their subscribers based on a three-part tariff package that consists of a fixed monthly fee, an amount of free minutes per month (bucket size), and an overage price that is charged for all minutes in excess of the stated number of free minutes. For a subscriber, with a constant known consumption rate, the choice of a package is straightforward. However, as in many other markets, mobile phone subscribers confront uncertainty regarding their usage. In this paper, we study the role of such usage uncertainty with respect to minutes on both a service provider’s package design and a subscriber’s choice of package. Specifically, we study how usage uncertainty affects a firm’s decisions about package size when it offers calling plans. In addition, from a subscriber’s perspective, we study the effect of uncertainty on whether a subscriber stays with a plan or upgrades to another. Our main results, (analytical and empirical) are that the effect of variation in usage on the subscriber’s optimal choice of free minute allotment and service package is conditional on whether the subscriber expects to use less than the entire bucket of minutes purchased. When this happens, the service provider’s optimal bucket size decreases as a function of the subscriber’s air-time variance. In addition, surprisingly, a subscriber’s usage variance, after controlling for mean usage, is actually a negative predictor of the subscriber choosing to upgrade.