Omid Rafieian
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Omid Rafieian has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Hema Yoganarasimhan
University of Washington, Seattle, Washington 98195
2 shared publications - Tommaso Bondi
Cornell University, New York, United States of America
2 shared publications - Yunfei (Jesse) Yao
The Chinese University of Hong Kong, Shatin, Hong Kong
2 shared publications Amitt Sharma
Vdo.ai, Vancouver, British Columbia V5Y 0E3, Canada
1 shared publication- Anuj Kapoor
Robert J. Trulaske, Sr. College of Business, University of Missouri, Columbia, Missouri 65211
1 shared publication - Mohammad Mosaffa
Cornell University, New York, NY, USA
1 shared publication
Latest Publications
- 29 October 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
The Problem of Ignoring the Ignorability Assumption in Advertising Measurement: A Matrix Completion Solution
This paper frames the overlap violation problem in online advertising auctions as a missing-data problem and proposes a matrix completion approach to correct the resulting bias in observational studies with algorithmic ad allocation.
- 29 October 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
The Problem of Ignoring the Ignorability Assumption in Advertising Measurement: A Matrix Completion Solution
This paper frames the overlap violation problem in online advertising auctions as a missing-data problem and proposes a matrix completion approach to correct the resulting bias in observational studies with algorithmic ad allocation.
- Top Cited Papers
- 1 February 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 72 (2) , 1389-1409
Privacy and Polarization: An Inference-Based Framework
Advances in behavioral targeting allow media firms to better monetize based on advertising. However, behavioral ad targeting requires consumer tracking, which has heightened privacy concerns among consumers and regulators. We examine how stricter privacy regulations that ban consumer tracking affect media firms’ content strategies and ideological positioning. We consider a model where media firms choose their ideological positioning and advertising, whereas ideologically heterogeneous consumers select their preferred content based on both their ideology and idiosyncratic taste shocks. We compare two salient informational environments: (1) behavioral ad targeting, where perfect inference about consumers is allowed, and (2) contextual ad targeting, where consumer tracking is banned due to privacy regulations, and media firms can only infer consumer types based on their media choice. We show that privacy regulations that ban behavioral ad targeting incentivize media firms to shift toward more extreme and polarizing positioning in order to draw better inferences about consumer types, even though the shift to more extreme ideological positions can hurt both demand and consumer welfare from content consumption. Compared with the monopoly case, competition increases firms’ inference motives and leads to more polarized content over a wider range of parameters due to an inferential complementarity effect arising from consumer self-selection. Our research uncovers a previously unexplored relationship between privacy and polarization, shedding light on the potential unintended consequences of privacy regulations in media markets. This paper was accepted by Dmitri Kuksov, marketing. Funding: This work was supported by the NET Institute. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mnsc.2024.06054 .
- 1 February 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 72 (2) , 1389-1409
Top Cited PapersPrivacy and Polarization: An Inference-Based Framework
Advances in behavioral targeting allow media firms to better monetize based on advertising. However, behavioral ad targeting requires consumer tracking, which has heightened privacy concerns among consumers and regulators. We examine how stricter privacy regulations that ban consumer tracking affect media firms’ content strategies and ideological positioning. We consider a model where media firms choose their ideological positioning and advertising, whereas ideologically heterogeneous consumers select their preferred content based on both their ideology and idiosyncratic taste shocks. We compare two salient informational environments: (1) behavioral ad targeting, where perfect inference about consumers is allowed, and (2) contextual ad targeting, where consumer tracking is banned due to privacy regulations, and media firms can only infer consumer types based on their media choice. We show that privacy regulations that ban behavioral ad targeting incentivize media firms to shift toward more extreme and polarizing positioning in order to draw better inferences about consumer types, even though the shift to more extreme ideological positions can hurt both demand and consumer welfare from content consumption. Compared with the monopoly case, competition increases firms’ inference motives and leads to more polarized content over a wider range of parameters due to an inferential complementarity effect arising from consumer self-selection. Our research uncovers a previously unexplored relationship between privacy and polarization, shedding light on the potential unintended consequences of privacy regulations in media markets. This paper was accepted by Dmitri Kuksov, marketing. Funding: This work was supported by the NET Institute. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mnsc.2024.06054 .
- Open Access
- 2 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1107-1108
Visual Polarization Measurement Using Counterfactual Image Generation
- 2 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1107-1108
Open AccessVisual Polarization Measurement Using Counterfactual Image Generation
- 1 March 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
- Vol. 44 (2) , 457-477
Multiobjective Personalization of Marketing Interventions
This paper develops multiobjective personalization algorithms and applies them to a video advertising experiment with a high degree of substitution between the outcomes of interest to demonstrate the value created by multiobjective personalization.
- 1 March 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
- Vol. 44 (2) , 457-477
Multiobjective Personalization of Marketing Interventions
This paper develops multiobjective personalization algorithms and applies them to a video advertising experiment with a high degree of substitution between the outcomes of interest to demonstrate the value created by multiobjective personalization.
- Open Access
- 8 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 206-207
Privacy and Polarization: An Inference-Based Framework
Digital publishers increasingly use advertising as a monetization strategy. At the core of ad-based monetization is behavioral ad targeting, which creates a sustainable revenue stream for publishers and keeps online content mostly free. However, behavioral ad targeting requires the collection and use of consumer-level data, which leads to privacy concerns among consumers. Although the main intent of privacy regulations is to safeguard consumer privacy, a consistent finding from past empirical research is that privacy regulation hurts digital publishers. Specifically, prior research suggests that the revenue loss due to privacy regulations is more pronounced for general interest (vs. specialized) publishers, who have greater uncertainty about their consumer types without consumer tracking. For example, in the absence of consumer tracking, a mainstream news website such as the New York Times - which attracts a highly heterogeneous pool of readers - has a harder time inferring consumer types and interests in order to show them relevant ads, compared to a niche, ideologically extreme website such as Infowars. The content-dependent impact of privacy regulation on media firms gives rise to an important question: does privacy regulation affect media firms' ideological positioning and content strategies? More specifically, what is the relationship between privacy and content polarization? To answer the above questions, we consider a model where media firms choose both their ideological positioning and advertising, while ideologically heterogeneous consumers select their preferred content based on both their ideology and idiosyncratic taste shocks. We compare two salient informational environments: (1) behavioral ad targeting, where perfect inference about consumers is allowed, and (2) contextual ad targeting, where consumer tracking is banned due to privacy regulations, and media firms can only make inference about consumer types based on their media choices. We show that, in both monopoly and duopoly contexts, privacy regulations that ban behavioral ad targeting incentivize media firms to shift toward more extreme and polarizing ideological positioning in order to draw better inferences about consumer types, even though the shift to more extreme ideological positions can hurt both demand and consumer welfare from content consumption. This incentive is stronger for media firms that generate lower quality content. Moreover, compared to the monopoly case, competition increases firms' inference motives and leads to more polarized content over a wider range of parameters due to an inferential complementarity effect arising from consumer self-selection. Our research uncovers a previously unexplored relationship between privacy and polarization: specifically, it goes against extensive media speculations and policy memos that cite hyper-personalization of online content and the lack of privacy regulation as a key contributor to the increased polarization over the past few decades. In our model, we provide new insights into this problem: it is the inability to personalize that leads firms to move to more extreme and polarizing ideological positioning to sharpen their inferences about their consumers. Put differently, in our paper, firms' polarizing strategies are motivated not by what firms know about consumers, but rather by what firms do not know and are trying to learn. Thus, privacy regulation can lead to higher content polarization compared to the case of perfect content personalization. A full version of this paper can be found at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4641822
- 8 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 206-207
Open AccessPrivacy and Polarization: An Inference-Based Framework
Digital publishers increasingly use advertising as a monetization strategy. At the core of ad-based monetization is behavioral ad targeting, which creates a sustainable revenue stream for publishers and keeps online content mostly free. However, behavioral ad targeting requires the collection and use of consumer-level data, which leads to privacy concerns among consumers. Although the main intent of privacy regulations is to safeguard consumer privacy, a consistent finding from past empirical research is that privacy regulation hurts digital publishers. Specifically, prior research suggests that the revenue loss due to privacy regulations is more pronounced for general interest (vs. specialized) publishers, who have greater uncertainty about their consumer types without consumer tracking. For example, in the absence of consumer tracking, a mainstream news website such as the New York Times - which attracts a highly heterogeneous pool of readers - has a harder time inferring consumer types and interests in order to show them relevant ads, compared to a niche, ideologically extreme website such as Infowars. The content-dependent impact of privacy regulation on media firms gives rise to an important question: does privacy regulation affect media firms' ideological positioning and content strategies? More specifically, what is the relationship between privacy and content polarization? To answer the above questions, we consider a model where media firms choose both their ideological positioning and advertising, while ideologically heterogeneous consumers select their preferred content based on both their ideology and idiosyncratic taste shocks. We compare two salient informational environments: (1) behavioral ad targeting, where perfect inference about consumers is allowed, and (2) contextual ad targeting, where consumer tracking is banned due to privacy regulations, and media firms can only make inference about consumer types based on their media choices. We show that, in both monopoly and duopoly contexts, privacy regulations that ban behavioral ad targeting incentivize media firms to shift toward more extreme and polarizing ideological positioning in order to draw better inferences about consumer types, even though the shift to more extreme ideological positions can hurt both demand and consumer welfare from content consumption. This incentive is stronger for media firms that generate lower quality content. Moreover, compared to the monopoly case, competition increases firms' inference motives and leads to more polarized content over a wider range of parameters due to an inferential complementarity effect arising from consumer self-selection. Our research uncovers a previously unexplored relationship between privacy and polarization: specifically, it goes against extensive media speculations and policy memos that cite hyper-personalization of online content and the lack of privacy regulation as a key contributor to the increased polarization over the past few decades. In our model, we provide new insights into this problem: it is the inability to personalize that leads firms to move to more extreme and polarizing ideological positioning to sharpen their inferences about their consumers. Put differently, in our paper, firms' polarizing strategies are motivated not by what firms know about consumers, but rather by what firms do not know and are trying to learn. Thus, privacy regulation can lead to higher content polarization compared to the case of perfect content personalization. A full version of this paper can be found at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4641822
- 1 September 2023
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
- Vol. 42 (5) , 910-933
Optimizing User Engagement Through Adaptive Ad Sequencing
This paper develops an offline reinforcement learning framework that identifies and evaluate the ad sequencing policy that optimizes user engagement.
- 1 September 2023
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
- Vol. 42 (5) , 910-933
Optimizing User Engagement Through Adaptive Ad Sequencing
This paper develops an offline reinforcement learning framework that identifies and evaluate the ad sequencing policy that optimizes user engagement.
- Top Cited Papers
- 1 March 2021
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
- Vol. 40 (2) , 193-218
Targeting and Privacy in Mobile Advertising
Mobile in-app advertising is now the dominant form of digital advertising. Although these ads have excellent user-tracking properties, they have raised concerns among privacy advocates. This has resulted in an ongoing debate on the value of different types of targeting information, the incentives of ad networks to engage in behavioral targeting, and the role of regulation. To answer these questions, we propose a unified modeling framework that consists of two components—a machine learning framework for targeting and an analytical auction model for examining market outcomes under counterfactual targeting regimes. We apply our framework to large-scale data from the leading in-app ad network of an Asian country. We find that an efficient targeting policy based on our machine learning framework improves the average click-through rate by 66.80% over the current system. These gains mainly stem from behavioral information compared with contextual information. Theoretical and empirical counterfactuals show that although total surplus grows with more granular targeting, the ad network’s revenues are nonmonotonic; that is, the most efficient targeting does not maximize ad network revenues. Rather, it is maximized when the ad network does not allow advertisers to engage in behavioral targeting. Our results suggest that ad networks may have economic incentives to preserve users’ privacy without external regulation.
- 1 March 2021
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
- Vol. 40 (2) , 193-218
Top Cited PapersTargeting and Privacy in Mobile Advertising
Mobile in-app advertising is now the dominant form of digital advertising. Although these ads have excellent user-tracking properties, they have raised concerns among privacy advocates. This has resulted in an ongoing debate on the value of different types of targeting information, the incentives of ad networks to engage in behavioral targeting, and the role of regulation. To answer these questions, we propose a unified modeling framework that consists of two components—a machine learning framework for targeting and an analytical auction model for examining market outcomes under counterfactual targeting regimes. We apply our framework to large-scale data from the leading in-app ad network of an Asian country. We find that an efficient targeting policy based on our machine learning framework improves the average click-through rate by 66.80% over the current system. These gains mainly stem from behavioral information compared with contextual information. Theoretical and empirical counterfactuals show that although total surplus grows with more granular targeting, the ad network’s revenues are nonmonotonic; that is, the most efficient targeting does not maximize ad network revenues. Rather, it is maximized when the ad network does not allow advertisers to engage in behavioral targeting. Our results suggest that ad networks may have economic incentives to preserve users’ privacy without external regulation.