Measuring Multichannel Advertising Response

Abstract

Advances in data collection have made it increasingly easy to collect information on advertising exposures. However, translating this seemingly rich data into measures of advertising response has proven difficult, largely because of concerns that advertisers target customers with a higher propensity to buy or increase advertising during periods of peak demand. We show how this problem can be addressed by studying a setting where a firm randomly held out customers from each campaign, creating a sequence of randomized field experiments that mitigates (many) potential endogeneity problems. Exploratory analysis of individual holdout experiments shows positive effects for both email and catalog; however, the estimated effect for any individual campaign is imprecise, because of the small size of the holdout. To pool data across campaigns, we develop a hierarchical Bayesian model for advertising response that allows us to account for individual differences in purchase propensity and marketing response. Building on the traditional ad-stock framework, we are able to estimate separate decay rates for each advertising medium, allowing us to predict channel-specific short- and long-term effects of advertising and use these predictions to inform marketing strategy. We find that catalogs have substantially longer-lasting impact on customer purchase than emails. We show how the model can be used to score and target individual customers based on their advertising responsiveness, and we find that targeting the most responsive customers increases the predicted returns on advertising by approximately 70% versus traditional recency, frequency, and monetary value–based targeting. This paper was accepted by Pradeep Chintagunta, marketing.

Informational Challenges in Omnichannel Marketing: Remedies and Future Research

“…Anderl et al (2016) propose a graph-based attribution model that maps the sequential nature of customer paths as first- and higher-order Markov walks and shows the idiosyncratic channel preferences (carryover) and interaction effects both within and across channel categories (spillover). Zantedeschi, Feit, and Bradlow (2017) develop a hierarchical Bayesian model for individual differences in purchase propensity and marketing response across channels, finding that catalogs have a substantially longer-lasting purchase impact on customer purchase than emails.…”

Inefficiencies in Digital Advertising Markets

“…Following are some questions that researchers could consider for the future and some brief thoughts on how to make progress on them: Advertisers can improve the usefulness of observational models. One strategy is to practice continuous experimentation across geographies, media, and groups of consumers, building more random variation into the data (Zantedeschi, Feit, and Bradlow 2017). For example, some digital advertising platforms offer targeting at the zip code level, which could enable greater statistical power than traditional market-level randomizations.…”

Charting the Path to Purchase Using Topic Models

“…Our model can be used to improve the quality of downstream experiments. Firms now routinely perform A/B testing to evaluate alternative decision options (e.g., price, promotion, advertising creative), and the effectiveness of an option often depends on a customer’s purchase stage (Hoban and Bucklin 2015; Zantedeschi, Feit, and Bradlow 2017). Because this stage is typically latent, observed or derived proxies are often used in empirical analyses.…”

“…The latent topics pursued by consumers and the dynamics of latent topics from time to time can serve as important indicators for marketing managers to track and evaluate, and the output can be incorporated into the firm’s automated marketing planning and allocation. Using clickstream data (Montgomery et al 2004; Trusov, Ma, and Jamal 2016) or designing experiments (Liu and Toubia 2018; Zantedeschi, Feit, and Bradlow 2017) can supplement our research with insights regarding consumers’ online browsing behaviors across multiple websites.…”