Download CV

contact

  • rafieian@uw.edu
  • Foster School of Business,
    University of Washington,
    347 Mackenzie Hall, Seattle, WA 98195

Omid Rafieian

  • Ph.D. Candidate in Marketing
  • University of Washington, Foster School of Business

Biography

I am a Ph.D. candidate in quantitative marketing at the Foster School of Business, University of Washington. My research interests broadly encompass topics related to digital marketing, mobile advertising, personalization, and privacy. I examine these topics through two complementary lenses – (1) how can we utilize the recent advancements in machine learning to create value in digital marketplaces, and (2) how can we use theory-driven structural frameworks to study the marketing and economic implications of such developments.

Research Interests

Substantive areas: digital marketing, mobile advertising, targeting, personalization, privacy, online auctions.

Methods: policy evaluation, structural models, machine learning, reinforcement learning, mechanism design, causal inference.

Education

2020

Doctor of Philosophy

in Marketing
University of Washington

2017

Master of Science in Business Administration

in Marketing
University of Washington

2015

Bachelor of Science

in Applied Mathematics
Sharif University of Technology

Job Market Papers

Rafieian, Omid, "Optimizing User Engagement through Adaptive Ad Sequencing."
(Job Market Paper I)

    Abstract: Mobile in-app advertising has grown exponentially in the last few years. In-app ads are often shown in a sequence of short-lived exposures for the duration of a user's stay in an app. The current state of both research and practice ignores the dynamics of ad sequencing and instead adopts a myopic framework to serve ads. In this paper, we propose a unified dynamic framework to adaptively sequence ads that comprises of two components – (1) a Markov Decision Process that captures the domain structure and incorporates inter-temporal trade-offs in ad interventions, and (2) an empirical framework that combines machine learning methods such as Extreme Gradient Boosting (XGBoost) with ideas from the causal inference literature to obtain counterfactual estimates of user behavior. We apply our framework to large-scale data from the leading in-app ad-network of an Asian country. We document significant gains from adopting a dynamic framework. We show that our forward-looking ad sequencing policy outperforms all the existing methods by comparing it to a series of benchmark policies often used in research and practice. Further, we demonstrate that these gains are heterogeneous across sessions: dynamic sequencing is most effective when users are new to the platform. Finally, we use a descriptive approach to explain the gains from adopting the dynamic framework.


Rafieian, Omid, "Revenue-Optimal Dynamic Auctions for Adaptive Ad Sequencing."
(Job Market Paper II)

    Abstract: Digital publishers often use real-time auctions to allocate their advertising inventory. These auctions are designed with the assumption that advertising exposures within a user's browsing or app-usage session are independent. Rafieian (2019) empirically documents the interdependence in the sequence of ads in mobile in-app advertising, and shows that dynamic sequencing of ads can improve the match between users and ads. In this paper, we examine the revenue gains from adopting a revenue-optimal dynamic auction. We propose a unified framework with two components – (1) a theoretical framework to derive the revenue-optimal dynamic auction that captures both advertisers' strategic bidding and users' ad response and app usage, and (2) an empirical framework that involves the structural estimation of advertisers' click valuations as well as personalized estimation of users' behavior using machine learning techniques. We apply our framework to large-scale data from the leading in-app ad-network of an Asian country. We document significant revenue gains from using the revenue-optimal dynamic auction compared to the revenue-optimal static auction. These gains stem from the improvement in the match between users and ads in the dynamic auction. The revenue-optimal dynamic auction also improves all key market outcomes, such as the total surplus, average advertisers' surplus, and market concentration.

Working Papers

Rafieian, Omid, and Yoganarasimhan, Hema, "Targeting and Privacy in Mobile Advertising."
Under Second Round Review at Marketing Science

    Abstract: Mobile in-app advertising is growing in popularity. While these ads have excellent user-tracking properties through mobile device IDs, 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 to contextual information. Theoretical and empirical counterfactuals show that while total surplus grows with more granular targeting, ad-network's revenues are non-monotonic, i.e., 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 incentives to preserve users' privacy without external regulation.


Rafieian, Omid, and Yoganarasimhan, Hema, "How Does Variety of Previous Ads Influence Consumer’s Ad Response?"

    Abstract: Mobile in-app advertising is now a major source of revenue for many app developers. In this paper, we focus on a unique aspect of in-app advertising – sequential ad placement. In this form of advertising, users are exposed to a sequence of potentially different ads within a session. This gives rise to a series of questions related to the effects of ad sequences. In particular, we are interested in the effects of variety of previous ads on user's clicking behavior on the next ad. We use data from the leading in-app ad-network from an Asian country to examine this question. A unique feature of our data is the use of probabilistic auction for ad placement that has created great variation in the sequence of ads users are exposed to within the session. Using this feature, we develop an identification strategy that allows us to obtain intent-to-treat estimates. We find that when exposed to a higher variety of previous ads, users are more likely to click on the next ad. We then explore the sources for the effects of variety and identify the sequential organization of exposures as a major source. This motivates us to develop a measure of sequential variety that capture variety of objects when presented in a sequence. Finally, we show the heterogeneity in the effects of variety across user's past history.

Work in Progress

Rafieian, Omid, "Geographical and Behavioral Information: Complements or Substitutes in Mobile Ad Targeting?"

Rafieian, Omid, "Benefits of Randomization in Online Ad Auctions."

Rafieian, Omid, "Value of User Identifiers in Mobile Ad Targeting."