Walter W. Zhang

Assistant Professor of Marketing

Contact Information

Primary Email:
wwz@wharton.upenn.edu
Office Phone:
(215) 898-9264

office Address:
744 Jon M. Huntsman Hall
3730 Walnut Street
University of Pennsylvania
Philadelphia, PA 19104

Research Interests: AI, Personalization, Targeting, Incentives, Platforms, Machine Learning, Causal Inference, Applied Optimal Transport

Links: Personal Website, CV

Overview

Walter W. Zhang is an Assistant Professor of Marketing at the Wharton School of the University of Pennsylvania. His research focuses on a central question in marketing: how should firms target and personalize their marketing mix under real-world constraints?

In his first line of research, he asks how firms can optimize their targeting strategy while addressing implementation and regulatory constraints. He quantifies the size of the constraints’ distortion on firm profits and designs solutions that leverage modern tools from economics and statistics to maximize profits while abiding by these constraints.

In his second line of research, he investigates how consumers respond to such targeting and personalization policies. He constructs and estimates structural models of consumer demand that incorporate the firm’s marketing mix variables and then uses these models to study counterfactual policies in welfare analyses.

Walter holds a PhD and an MBA from the University of Chicago Booth School of Business along with bachelor’s degrees in Physics and Economics from the University of Chicago.

For more information, please visit Walter’s website: walterwzhang.github.io

Research

Lin Fei, Daniel M. Bartels, Walter W. Zhang (Draft), Consumers’ Mental Representation of Expenditures: Implications for Spending and Saving Decisions. Abstract
People’s mental representation of expenditures is crucial to how they budget. We propose that much like how people represent natural kinds (e.g., animals and plants), people represent expenditures in a hierarchical taxonomy. Across seven studies, supported by six norming studies and three pilots, we found evidence of a hierarchical representation of expenditures. We first recover people’s mental representations using a successive pile-sort method that asks people to form hierarchies of categories with common expenditures (e.g., rent, dining out, etc.). We found that there is consensus in people’s hierarchical representations of expenditures and that their representations are relatively stable over time. Further, we found that people’s adjustment in their spending behavior can be predicted by the distance between items in their representation. Specifically, when people overspent on an item, they were more likely to spontaneously adjust spending for items closer in representation than those further away. We examine this spontaneous adjustment behavior using both lab studies and field data with 6.5 million grocery shopping trips over twelve years. The findings highlight the connection between mental representation and consumer behavior, and they emphasize the importance of studying concepts and categories in the context of consumption.
Walter W. Zhang (Draft), Optimal Comprehensible Targeting. Abstract
Developments in machine learning and big data allow firms to fully personalize and target their marketing mix. However, data and privacy regulations, such as those in the European Union (GDPR), incorporate a “right to explanation”, which is fulfilled when targeting policies are comprehensible to customers. This paper provides a framework for firms to navigate right-to-explanation laws. First, I introduce a new method called Policy DNN, which combines policy learning and deep neural networks, to form a profit-maximizing black box benchmark and provide theoretical guarantees on its performance. In contrast to prior approaches that use a two-step method of estimating treatment effects before assigning individuals their treatment group, Policy DNN directly estimates treatment assignment, which improves efficiency. Second, I construct a class of comprehensible targeting policies that is represented by a sentence. Third, I show how to optimize over this class of policies to find the profit-maximizing comprehensible policy. I demonstrate that it is optimal to estimate the comprehensible policy directly from the data, rather than projecting down the black box policy into a comprehensible policy. Finally, I apply my framework empirically in the context of price promotions for a durable goods retailer using data from a field experiment. I quantify the cost of explanation, which I define as the difference in expected profits between the optimal black box and comprehensible targeting policies. The comprehensible targeting policy reduces profits by 7% or 22 cents per customer when compared to the black box benchmark.
Günter Hitsch, Sanjog Misra, Walter W. Zhang (2024), Heterogeneous treatment effects and optimal targeting policy evaluation, Quantitative Marketing and Economics, Volume 22 (2024), pp. 115-168. Abstract
We present a general framework to target customers using optimal targeting policies, and we document the profit differences from alternative estimates of the optimal targeting policies. Two foundations of the framework are conditional average treatment effects (CATEs) and off-policy evaluation using data with randomized targeting. This policy evaluation approach allows us to evaluate an arbitrary number of different targeting policies using only one randomized data set and thus provides large cost advantages over conducting a corresponding number of field experiments. We use different CATE estimation methods to construct and compare alternative targeting policies. Our particular focus is on the distinction between indirect and direct methods. The indirect methods predict the CATEs using a conditional expectation function estimated on outcome levels, whereas the direct methods specifically predict the treatment effects of targeting. We introduce a new direct estimation method called treatment effect projection (TEP). The TEP is a non-parametric CATE estimator that we regularize using a transformed outcome loss which, in expectation, is identical to a loss that we could construct if the individual treatment effects were observed. The empirical application is to a catalog mailing with a high-dimensional set of customer features. We document the profits of the estimated policies using data from two campaigns conducted one year apart, which allows us to assess the transportability of the predictions to a campaign implemented one year after collecting the training data. All estimates of the optimal targeting policies yield larger profits than uniform policies that target none or all customers. Further, there are significant profit differences across the methods, with the direct estimation methods yielding substantially larger economic value than the indirect methods.
Walter W. Zhang and Sanjog Misra (Draft), Coarse Personalization. Abstract
Advances in estimating heterogeneous treatment effects enable firms to personalize marketing mix elements and target individuals at an unmatched level of granularity, but feasibility constraints limit such personalization. In practice, firms choose which unique treatments to offer and which individuals to offer these treatments with the goal of maximizing profits: we call this the coarse personalization problem. We propose a two-step solution that makes segmentation and targeting decisions in concert. First, the firm personalizes by estimating conditional average treatment effects. Second, the firm discretizes by utilizing treatment effects to choose which unique treatments to offer and who to assign to these treatments. We show that a combination of available machine learning tools for estimating heterogeneous treatment effects and a novel application of optimal transport methods provides a viable and efficient solution. With data from a large-scale field experiment for promotions management, we find that our methodology outperforms extant approaches that segment on consumer characteristics or preferences and those that only search over a prespecified grid. Using our procedure, the firm recoups over 99.5% of its expected incremental profits under fully granular personalization while offering only five unique treatments. We conclude by discussing how coarse personalization arises in other domains.

Teaching

Past Courses

MKTG1010 - Intro To Marketing
The objective of this course is to introduce students to the concepts, analyses, and activities that comprise marketing management, and to provide practice in assessing and solving marketing problems. The course is also a foundation for advanced electives in Marketing as well as other business/social disciplines. Topics include marketing strategy, customer behavior, segmentation, customer lifetime value, branding, market research, product lifecycle strategies, pricing, go-to-market strategies, promotion, and marketing ethics.
MKTG1018 - Intro To Marketing
The objective of this course is to introduce students to the concepts, analyses, and activities that comprise marketing management, and to provide practice in assessing and solving marketing problems. The course is also a foundation for advanced electives in Marketing as well as other business/social disciplines. Topics include marketing strategy, customer behavior, segmentation, customer lifetime value, branding, market research, product lifecycle strategies, pricing, go-to-market strategies, promotion, and marketing ethics. (This is the honors section of MKTG 1010 open only to Joseph Wharton Scholars).

Activity

Latest Research

Lin Fei, Daniel M. Bartels, Walter W. Zhang (Draft), Consumers’ Mental Representation of Expenditures: Implications for Spending and Saving Decisions.

All Research