Peter Ebbes

Firms increasingly rely on A/B testing to evaluate marketing strategies, yet most experiments are analyzed in isolation, limiting insight into why effectiveness varies and how repeated exposure shapes outcomes. We develop a hierarchical Bayesian framework that jointly analyzes randomized marketing interventions to decompose treatment effect heterogeneity into three components: customer responsiveness, campaign design, and contextual timing. The model is scalable, incorporates both observed and unobserved variation, and links estimation directly to policy evaluation with full uncertainty quantification. Using data from large-scale field experiments involving nearly half a million customers, we document three main findings. First, unobserved customer-level heterogeneity accounts for the majority of variation in treatment effects, explaining roughly two-thirds of the total dispersion, considerably exceeding the contribution of campaign design or timing. Second, we find strong evidence of intervention fatigue: responsiveness declines with repeated exposure and recovers only slowly. Third, in a held-out policy evaluation, a model-based targeting strategy improves revenue on average while substantially reducing the share of customers targeted, with gains reaching over 10% in settings where experimental variation is most informative. Approximately 80% of these gains are driven by personalization based on latent customer responsiveness rather than observable characteristics. Together, these results demonstrate how repeated experimentation can be leveraged to explain variation in marketing effectiveness and to support scalable targeting and personalization.

This chapter takes networks as the default lens for interorganizational research (IOR) in marketing, guiding readers from theory to practice. The chapter has four main contributions. First, we provide an overview of common pitfalls, suggested fixes, and a checklist to avoid them. Second, we offer a portable theories → constructs → metrics → data map that links key theoretical mechanisms to appropriate metrics and data requirements. We organize metrics in a macro-meso-micro scaffold, distinguishing macro-level network characteristics (e.g., degree distribution), meso-level characteristics (e.g., communities), and micro-level characteristics (e.g., network-position metrics). Third, we propose a two-by-two typology of empirical designs, unit of inference (node vs. tie) by dominant mechanism (ego-vs. structure-driven), that specifies characteristic covariates, identification needs, and statistical methods. Lastly, we offer a roadmap for future IOR research in marketing strategy. Throughout the chapter, we use a running boardinterlock example to make concrete the move from director-firm affiliations (two-mode) to firmfirm ties (one-mode), metric computation and interpretation, community detection considerations, and modeling board-interlock tie formation using Exponential Random Graph Models. Taken together, the chapter offers the reader a guide from foundations to data structures, metrics, and formation models in complex IOR settings.

An important challenge for many firms is to identify the life transitions of its customers, such as job searching, expecting a child, or purchasing a home. Inferring such transitions, which are generally unobserved to the firm, can offer the firms opportunities to be more relevant to their customers. In this paper, we demonstrate how a social network platform can leverage its longitudinal user data to identify which of its users are likely to be job seekers. Identifying job seekers is at the heart of the business model of professional social network platforms. Our proposed approach builds on the hidden Markov model (HMM) framework to recover the latent state of job search from noisy signals obtained from social network activity data. Specifically, we use the latent states of the HMM to fuse cross-sectional survey responses to a job-seeking status question with longitudinal user activity data, resulting in a partially HMM. Thus, in some time periods, and for some users, we observe a direct measure of the true job-seeking status. We demonstrate that the proposed model can predict not only which users are likely to be job seeking at any point in time but also what activities on the platform are associated with job search and how long the users have been job seeking. Furthermore, we find that targeting job seekers based on our proposed approach can lead to a 29% increase in profits of a targeting campaign relative to the approach that was used by the social network platform.This paper was accepted by Juanjuan Zhang, marketing.

Consumers often arrive at the point of purchase with a particular product to purchase in mind and must search for this target product in the product display. Drawing on visual crowding theory, we suggest that an individual’s ability to find a target product in a display varies according to the embeddedness of the target, which depends on the locations of the target and non-target products in the display. We refer to the visual impact of non-target products on target products that results in an inability to differentiate and recognize the target product as “target visual crowdedness (TVC).” Across five experimental studies, we manipulate TVC and show that as TVC increases the duration of the visual search increases. Eye-tracking data provide evidence of how TVC affects attention and information processing in target product search tasks. We develop a visual crowdedness metric that quantifies TVC and can be used by researchers and practitioners to estimate the visual search effort needed to find a product in a display.

Customer relationship management (CRM) campaigns have traditionally focused on maximizing the profitability of the targeted customers. The authors demonstrate that in business settings characterized by network externalities, a CRM campaign that is aimed at changing the behavior of specific customers propagates through the social network, thereby also affecting the behavior of nontargeted customers. Using a randomized field experiment involving nearly 6,000 customers of a mobile telecommunication provider, they find that the social connections of targeted customers increase their consumption and become less likely to churn, due to a campaign that was neither targeted at them nor offered them any direct incentives. The authors estimate a social multiplier of 1.28. That is, the effect of the campaign on first-degree connections of targeted customers is 28% of the effect of the campaign on the targeted customers. By further leveraging the randomized experimental design, the authors show that, consistent with a network externality account, the increase in activity among the nontargeted but connected customers is driven by the increase in communication between the targeted customers and their connections, making the local network of the nontargeted customers more valuable. These findings suggest that in targeting CRM marketing campaigns, firms should consider not only the profitability of the targeted customer but also the potential spillover of the campaign to nontargeted but connected customers.

Marketing academics and practitioners alike remain unconvinced about the chief marketing officer's (CMO's) performance implications. Whereas some studies propose that firms benefit financially from having a CMO in the C-suite, other studies conclude that the CMO has little or no effect on firm performance. Accordingly, there have been strong calls for additional academic research regarding the CMO's performance implications. In response to these calls, the authors employ model specifications with varying identifying assumptions (i.e., rich data models, unobserved effects models, instrumental variable models, and panel internal instruments models) and use data from up to 155 publicly traded firms over a 12-year period (2000–2011) to find that firms can indeed expect to benefit financially from having a CMO at the strategy table. Specifically, their findings suggest that the performance (measured in terms of Tobin's q) of the sample firms that employ a CMO is, on average, approximately 15% greater than that of the sample firms that do not employ a CMO. This result is robust to the type of model specification used. Marketing academics and practitioners should find the results intriguing given the existing uncertainty surrounding the CMO's performance implications. The study also contributes to the methodology literature by collating diverse empirical model specifications that can be used to model causal effects with observational data into a coherent and comprehensive framework.