Xuying Zhao
Research Subjects
Xuying Zhao has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
- Kathryn E. Stecke
University of Texas at Dallas
4 shared publications Kathryn E. Stecke
University of Texas at Dallas - Naveen Jindal School of Management
3 shared publications- Ashutosh Prasad
University of Texas at Dallas
2 shared publications - Ashutosh Prasad
School of Business, University of California Riverside, Riverside, California 92521
2 shared publications - Daewon Sun
Department of Management Mendoza College of Business University of Notre Dame Notre Dame IN 46556
2 shared publications - Zhan Pang
Independent
2 shared publications
Latest Publications
- 16 September 2026
- Published by SAGE Publications in Production and Operations Management
Let clickstream talk: A graph neural network approach to sales forecasting
: This article proposes ForecastClickGraph, a deep learning framework that extracts cross-product relationships and demand information from clickstream data for probabilistic sales forecasting. ForecastClickGraph models stages in consumer shopping journeys as a dynamic graph, where nodes represent individual product-stage combinations, such as product A viewed or added to cart, and directed edges capture the transitions between stages. A customized graph neural network learns product representations that incorporate both product-level temporal patterns and cross-product associations. Another graph-based module further detects demand spikes using early signals from related products. ForecastClickGraph cross-learns demand patterns across large-scale products and estimates demand distributions through quantile regression. Extensive experiments on a real-world dataset show that ForecastClickGraph outperforms state-of-the-art benchmark models by 10%–28% in forecast accuracy, with particularly strong performance in predicting promotional sales bursts. It also yields superior probabilistic forecasts with substantially lower quantile loss and improved calibration relative to the benchmarks.
- 16 September 2026
- Published by SAGE Publications in Production and Operations Management
Let clickstream talk: A graph neural network approach to sales forecasting
: This article proposes ForecastClickGraph, a deep learning framework that extracts cross-product relationships and demand information from clickstream data for probabilistic sales forecasting. ForecastClickGraph models stages in consumer shopping journeys as a dynamic graph, where nodes represent individual product-stage combinations, such as product A viewed or added to cart, and directed edges capture the transitions between stages. A customized graph neural network learns product representations that incorporate both product-level temporal patterns and cross-product associations. Another graph-based module further detects demand spikes using early signals from related products. ForecastClickGraph cross-learns demand patterns across large-scale products and estimates demand distributions through quantile regression. Extensive experiments on a real-world dataset show that ForecastClickGraph outperforms state-of-the-art benchmark models by 10%–28% in forecast accuracy, with particularly strong performance in predicting promotional sales bursts. It also yields superior probabilistic forecasts with substantially lower quantile loss and improved calibration relative to the benchmarks.
- Top Cited Papers
- 1 August 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 71 (8) , 6816-6835
A Model of Shoppertainment Live Streaming
“Shoppertainment” live streaming, a blend of entertainment and product selling, is becoming increasingly popular. In shoppertainment sessions, streamers face bandwidth constraints so balance between providing product information and entertainment (such as singing, dancing, and storytelling). We conceptualize a shoppertainment live streamer as a novel type of online retail platform that maximizes total commissions from selling a range of manufacturer products. Our study investigates how streamers can leverage entertainment bandwidth as a long-term strategy to influence manufacturers’ pricing decisions and consumers’ attendance, as well as purchasing behaviors. Our model reveals several intriguing findings. First, we find that higher entertainment bandwidth encourages more manufacturers to adopt a demand-oriented, low-pricing strategy to attract buyers rather than a margin-oriented, high-pricing strategy that appeals only to buyers who perceive a high likelihood of finding a suitable product. Second, we observe that an increase in entertainment bandwidth initially boosts consumer traffic to shoppertainment sessions, but this trend may reverse once the bandwidth becomes excessively high, indicating an inverted U-shaped relationship. Conversely, the streamer’s expected profit per attending consumer first decreases and then increases with higher entertainment bandwidth, forming a U-shaped relationship. Third, our results suggest that a commission-maximizing streamer facing bandwidth constraints should not exclusively focus on providing product information. Instead, it is advantageous to allocate significant bandwidth to entertainment, even with limited entertaining capability, and potentially beneficial to dedicate all bandwidth to entertainment if the capability is substantial. Lastly, we demonstrate that the streamer’s quality threshold moderates the impact of entertainment bandwidth on these dynamics. This paper was accepted by David Simchi-Levi, information systems. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mnsc.2023.01724 .
- 1 August 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 71 (8) , 6816-6835
Top Cited PapersA Model of Shoppertainment Live Streaming
“Shoppertainment” live streaming, a blend of entertainment and product selling, is becoming increasingly popular. In shoppertainment sessions, streamers face bandwidth constraints so balance between providing product information and entertainment (such as singing, dancing, and storytelling). We conceptualize a shoppertainment live streamer as a novel type of online retail platform that maximizes total commissions from selling a range of manufacturer products. Our study investigates how streamers can leverage entertainment bandwidth as a long-term strategy to influence manufacturers’ pricing decisions and consumers’ attendance, as well as purchasing behaviors. Our model reveals several intriguing findings. First, we find that higher entertainment bandwidth encourages more manufacturers to adopt a demand-oriented, low-pricing strategy to attract buyers rather than a margin-oriented, high-pricing strategy that appeals only to buyers who perceive a high likelihood of finding a suitable product. Second, we observe that an increase in entertainment bandwidth initially boosts consumer traffic to shoppertainment sessions, but this trend may reverse once the bandwidth becomes excessively high, indicating an inverted U-shaped relationship. Conversely, the streamer’s expected profit per attending consumer first decreases and then increases with higher entertainment bandwidth, forming a U-shaped relationship. Third, our results suggest that a commission-maximizing streamer facing bandwidth constraints should not exclusively focus on providing product information. Instead, it is advantageous to allocate significant bandwidth to entertainment, even with limited entertaining capability, and potentially beneficial to dedicate all bandwidth to entertainment if the capability is substantial. Lastly, we demonstrate that the streamer’s quality threshold moderates the impact of entertainment bandwidth on these dynamics. This paper was accepted by David Simchi-Levi, information systems. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mnsc.2023.01724 .
- 1 March 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 71 (3) , 2544-2564
Selling Bonus Actions in Video Games
In the mobile video games industry, a common in-app purchase is for additional “moves” or “time” in single-player puzzle games. We call these in-app purchases bonus actions. In some games, bonus actions can only be purchased in advance of attempting a level of the game (pure advance sales (PAS)), yet in other games, bonus actions can only be purchased in a “spot” market that appears when an initial attempt to pass the level fails (pure spot sales). Some games offer both advance and spot purchases (hybrid advance sales). This paper studies these selling strategies for bonus actions in video games. Such a question is novel to in-app tools selling in video games, and it cannot be answered by previous advance selling studies focusing on end goods. We model the selling of bonus actions as a stochastic extensive form game. We show how the distribution of skill among players (i.e., their inherent ability to pass the level) and the inherent randomness of the game influence selling strategies. For casual games, where low-skill players have a sufficiently high probability of success in each attempt, if the proportion of high-skill players is either sufficiently large or sufficiently small, firms should adopt PAS and shut down the “spot” market. Furthermore, the player welfare-maximizing selling strategy is to sell only in the spot market. Hence, no “win-win” strategy exists for casual games. However, PAS can be a win-win for hardcore games, where low-skill players have a sufficiently low success probability for each attempt. This paper was accepted by Hemant Bhargava, information systems. Funding: C. T. Ryan received funding from NSERC Discovery [Grant RGPIN-2020-06488]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mnsc.2022.02348 .
- 1 March 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 71 (3) , 2544-2564
Selling Bonus Actions in Video Games
In the mobile video games industry, a common in-app purchase is for additional “moves” or “time” in single-player puzzle games. We call these in-app purchases bonus actions. In some games, bonus actions can only be purchased in advance of attempting a level of the game (pure advance sales (PAS)), yet in other games, bonus actions can only be purchased in a “spot” market that appears when an initial attempt to pass the level fails (pure spot sales). Some games offer both advance and spot purchases (hybrid advance sales). This paper studies these selling strategies for bonus actions in video games. Such a question is novel to in-app tools selling in video games, and it cannot be answered by previous advance selling studies focusing on end goods. We model the selling of bonus actions as a stochastic extensive form game. We show how the distribution of skill among players (i.e., their inherent ability to pass the level) and the inherent randomness of the game influence selling strategies. For casual games, where low-skill players have a sufficiently high probability of success in each attempt, if the proportion of high-skill players is either sufficiently large or sufficiently small, firms should adopt PAS and shut down the “spot” market. Furthermore, the player welfare-maximizing selling strategy is to sell only in the spot market. Hence, no “win-win” strategy exists for casual games. However, PAS can be a win-win for hardcore games, where low-skill players have a sufficiently low success probability for each attempt. This paper was accepted by Hemant Bhargava, information systems. Funding: C. T. Ryan received funding from NSERC Discovery [Grant RGPIN-2020-06488]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mnsc.2022.02348 .
- 1 December 2024
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Information Systems Research
- Vol. 35 (4) , 1785-1801
Content Length Limit: How Does It Matter for a Consumer-to-Consumer Media Platform?
Our study is inspired by the rapid growth of consumer-to-consumer (C2C) media platforms such as TikTok. There are three key findings. First, we show that when content pieces on the platform are longer, viewers set a higher standard of match value in selecting content to view, leading to a lower click-through rate of contributed content on the platform. This finding suggests that a tight limit on content length increases click-through rate. Second, we show that extended content length on the platform first enhances platform performance but then hurts its performance, following an inverted U-shape curve. This pattern holds true for short-term performance measured by viewer traffic and total viewing time, as well as for long-term performance measured by total consumer surplus. This finding suggests the existence of an optimal content length. Third, we find that the optimal content length maximizing viewer traffic is smaller than the one maximizing total viewing time, which is further smaller than the one maximizing consumer surplus. As such, a platform that switches the strategic focus from short-term advertising revenue to long-term growth will benefit from extending the content length limit.
- 1 December 2024
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Information Systems Research
- Vol. 35 (4) , 1785-1801
Content Length Limit: How Does It Matter for a Consumer-to-Consumer Media Platform?
Our study is inspired by the rapid growth of consumer-to-consumer (C2C) media platforms such as TikTok. There are three key findings. First, we show that when content pieces on the platform are longer, viewers set a higher standard of match value in selecting content to view, leading to a lower click-through rate of contributed content on the platform. This finding suggests that a tight limit on content length increases click-through rate. Second, we show that extended content length on the platform first enhances platform performance but then hurts its performance, following an inverted U-shape curve. This pattern holds true for short-term performance measured by viewer traffic and total viewing time, as well as for long-term performance measured by total consumer surplus. This finding suggests the existence of an optimal content length. Third, we find that the optimal content length maximizing viewer traffic is smaller than the one maximizing total viewing time, which is further smaller than the one maximizing consumer surplus. As such, a platform that switches the strategic focus from short-term advertising revenue to long-term growth will benefit from extending the content length limit.
- 1 January 2021
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Manufacturing & Service Operations Management
- Vol. 23 (1) , 123-138
Preorder Price Guarantee in e-Commerce
Problem definition: With the development of the internet and e-commerce, retailers often offer preorders for new, to-be-released products. To encourage preorders, retailers such as Amazon offer preorder price guarantee (PG). That is, if the product price drops before or on the release date, preorder consumers automatically receive a refund for the difference between the preorder price and the new price. Should retailers offer PG for preorders? If so, how should they decide the prices and inventory level? Academic/practical relevance: Both advance selling and price matching have been observed in practice and studied in literature. However, the combination of these two is a new phenomenon in practice and receives very limited attention in academic. So our study is closely related to both academic and practical communities. Methodology: Dynamic pricing and game theory. Results: We find that, if preorder demand uncertainty is high, a firm should adopt PG in advance selling. If preorder demand uncertainty is low, then a firm should adopt PG if and only if the percentage of high-valuation consumers is high. Furthermore, we find that a firm's optimal profit under PG monotonically increases in preorder demand uncertainty while the firm's optimal profit without PG stays unchanged. That is, PG enables a firm to profit from preorder demand uncertainty. In addition, we show that price commitment is dominated by dynamic pricing when the retailer can optimally decide whether to offer PG under dynamic pricing. We also demonstrate that a retailer should sell in advance if a product's marginal cost is less than a certain threshold, which is higher than the traditional threshold in the advance selling literature without consideration of PG. Managerial implications: Our results provide guidance for retailers on whether or not they should offer PG and how to design the prices under PG.
- 1 January 2021
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Manufacturing & Service Operations Management
- Vol. 23 (1) , 123-138
Preorder Price Guarantee in e-Commerce
Problem definition: With the development of the internet and e-commerce, retailers often offer preorders for new, to-be-released products. To encourage preorders, retailers such as Amazon offer preorder price guarantee (PG). That is, if the product price drops before or on the release date, preorder consumers automatically receive a refund for the difference between the preorder price and the new price. Should retailers offer PG for preorders? If so, how should they decide the prices and inventory level? Academic/practical relevance: Both advance selling and price matching have been observed in practice and studied in literature. However, the combination of these two is a new phenomenon in practice and receives very limited attention in academic. So our study is closely related to both academic and practical communities. Methodology: Dynamic pricing and game theory. Results: We find that, if preorder demand uncertainty is high, a firm should adopt PG in advance selling. If preorder demand uncertainty is low, then a firm should adopt PG if and only if the percentage of high-valuation consumers is high. Furthermore, we find that a firm's optimal profit under PG monotonically increases in preorder demand uncertainty while the firm's optimal profit without PG stays unchanged. That is, PG enables a firm to profit from preorder demand uncertainty. In addition, we show that price commitment is dominated by dynamic pricing when the retailer can optimally decide whether to offer PG under dynamic pricing. We also demonstrate that a retailer should sell in advance if a product's marginal cost is less than a certain threshold, which is higher than the traditional threshold in the advance selling literature without consideration of PG. Managerial implications: Our results provide guidance for retailers on whether or not they should offer PG and how to design the prices under PG.
- Open Access
- 27 November 2020
- Published by Elsevier in SSRN Electronic Journal
Selling New Products with Social Learning: The Role of Online Product Reviews
This paper studies the impact of consumers' social learning from the reviews of early adopters of a new product. We consider a two-period setting with an introduction (penetration) period followed by the product maturing period. Early buyers make purchase decisions in the introduction period under significant product valuation uncertainty and future price uncertainty and would like post reviews online to share their experiences about the product. The online product reviews posted by early adopters play an important role in shaping following consumers' product valuations and further influence their purchasing decisions. While the majority of research on social learning focuses on the impact of social learning on the consumer decisions and firm pricing in the maturing period, we find an indirect impact of electronic word-of-mouth (EWOM) on the strategic early buyers' purchase decisions and firm pricing in the introduction period. When early buyers make purchase decisions, they need to forecast the future second period price, estimate the expected utility from waiting, and compare it with the expected utility from buying early. We find that social learning increases the uncertainty in future price, which depends on reviews and therefore unknown, for early buyers in the introduction period. This price-uncertainty-increasing effect may balk early buyers from adopting in the introduction period and thus impact a firm's profit. Driven by this effect, we find that under social learning negative reviews may benefit a firm while positive reviews can also hurt a firm.
- 27 November 2020
- Published by Elsevier in SSRN Electronic Journal
Open AccessSelling New Products with Social Learning: The Role of Online Product Reviews
This paper studies the impact of consumers' social learning from the reviews of early adopters of a new product. We consider a two-period setting with an introduction (penetration) period followed by the product maturing period. Early buyers make purchase decisions in the introduction period under significant product valuation uncertainty and future price uncertainty and would like post reviews online to share their experiences about the product. The online product reviews posted by early adopters play an important role in shaping following consumers' product valuations and further influence their purchasing decisions. While the majority of research on social learning focuses on the impact of social learning on the consumer decisions and firm pricing in the maturing period, we find an indirect impact of electronic word-of-mouth (EWOM) on the strategic early buyers' purchase decisions and firm pricing in the introduction period. When early buyers make purchase decisions, they need to forecast the future second period price, estimate the expected utility from waiting, and compare it with the expected utility from buying early. We find that social learning increases the uncertainty in future price, which depends on reviews and therefore unknown, for early buyers in the introduction period. This price-uncertainty-increasing effect may balk early buyers from adopting in the introduction period and thus impact a firm's profit. Driven by this effect, we find that under social learning negative reviews may benefit a firm while positive reviews can also hurt a firm.
- Open Access
- 25 April 2017
- Published by Elsevier in SSRN Electronic Journal
The Role of Expectation-Reality Discrepancy in Mobile Service Contracts
Mobile carriers used to enforce service contracts on consumers. However, recently major mobile carriers have eliminated contracts. We investigate carriers' contract and marketing strategies. We identify Expectation-Reality Discrepancy (ERD) as a key determinant. A carrier's ERD is defined as consumers' ex-ante expected valuation minus their ex-post realized valuation of the carrier's service. Main findings include: Carriers' contract strategies critically depend on their ERDs rather than the true service valuations. A carrier with a higher ERD is more likely to enforce contracts, regardless of whether the true service valuation is higher than that of her competitor. Carriers should enforce contracts only when they have positive ERDs. When both carriers have positive ERDs, cross switching occurs, resulting in that some consumers located closer to a carrier switch to the other carrier and vice versa. Contracts have the competition-intensifying effect. When carriers enforce contracts, their competition on promoting consumer expectations is intensified, leading to higher ERDs with contracts than without contracts. Finally, both contracts and positive ERD marketing strategies hurt consumer welfare.
- 25 April 2017
- Published by Elsevier in SSRN Electronic Journal
Open AccessThe Role of Expectation-Reality Discrepancy in Mobile Service Contracts
Mobile carriers used to enforce service contracts on consumers. However, recently major mobile carriers have eliminated contracts. We investigate carriers' contract and marketing strategies. We identify Expectation-Reality Discrepancy (ERD) as a key determinant. A carrier's ERD is defined as consumers' ex-ante expected valuation minus their ex-post realized valuation of the carrier's service. Main findings include: Carriers' contract strategies critically depend on their ERDs rather than the true service valuations. A carrier with a higher ERD is more likely to enforce contracts, regardless of whether the true service valuation is higher than that of her competitor. Carriers should enforce contracts only when they have positive ERDs. When both carriers have positive ERDs, cross switching occurs, resulting in that some consumers located closer to a carrier switch to the other carrier and vice versa. Contracts have the competition-intensifying effect. When carriers enforce contracts, their competition on promoting consumer expectations is intensified, leading to higher ERDs with contracts than without contracts. Finally, both contracts and positive ERD marketing strategies hurt consumer welfare.
- 1 June 2016
- Published by SAGE Publications in Production and Operations Management
- Vol. 25 (6) , 1073-1087
When Does a Retailer's Advance Selling Capability Benefit Manufacturer, Retailer, or Both?
Advance selling (AS) from a retailer to consumers is commonly observed in practice. With an AS capability, a retailer has the option to sell in advance or not. Having the AS option seems to increase flexibility and thus profit for a retailer. However, we show that the AS option can hurt the retailer's profit as well as supply chain performance. We identify two thresholds for a product's marginal production cost. A retailer's AS option benefits both the manufacturer and retailer when the marginal production cost is high, that is, above both thresholds. It benefits the manufacturer but hurts the retailer when the marginal production cost is moderate, that is, between the two thresholds. The result is ambiguous when the marginal production cost is low, that is, below both thresholds. We find that consumer valuation uncertainty under AS is the key driving force for the surprising result that having the retailer's AS option can hurt the retailer. When compared to the scenario where the retailer does not have the AS option, we find that the manufacturer's optimal wholesale price weakly decreases under the retailer's AS option if the marginal production cost is high. The statement is reversed if the marginal production cost is moderate or low.
- 1 June 2016
- Published by SAGE Publications in Production and Operations Management
- Vol. 25 (6) , 1073-1087
When Does a Retailer's Advance Selling Capability Benefit Manufacturer, Retailer, or Both?
Advance selling (AS) from a retailer to consumers is commonly observed in practice. With an AS capability, a retailer has the option to sell in advance or not. Having the AS option seems to increase flexibility and thus profit for a retailer. However, we show that the AS option can hurt the retailer's profit as well as supply chain performance. We identify two thresholds for a product's marginal production cost. A retailer's AS option benefits both the manufacturer and retailer when the marginal production cost is high, that is, above both thresholds. It benefits the manufacturer but hurts the retailer when the marginal production cost is moderate, that is, between the two thresholds. The result is ambiguous when the marginal production cost is low, that is, below both thresholds. We find that consumer valuation uncertainty under AS is the key driving force for the surprising result that having the retailer's AS option can hurt the retailer. When compared to the scenario where the retailer does not have the AS option, we find that the manufacturer's optimal wholesale price weakly decreases under the retailer's AS option if the marginal production cost is high. The statement is reversed if the marginal production cost is moderate or low.
- 23 September 2015
- Published by Wiley in Decision Sciences
- Vol. 47 (1) , 125-156
Newsvendor Problems with Demand Shocks and Unknown Demand Distributions
In today's competitive market, demand volume and even the underlying demand distribution can change quickly for a newsvendor seller. We refer to sudden changes in demand distribution as demand shocks. When a newsvendor seller has limited demand distribution information and also experiences underlying demand shocks, the majority of existing methods for newsvendor problems may not work well since they either require demand distribution information or assume stationary demand distribution. We present a new, robust, and effective machine learning algorithm for newsvendor problems with demand shocks but without any demand distribution information. The algorithm needs only an approximate estimate of the lower and upper bounds of demand range; no other knowledge such as demand mean, variance, or distribution type is necessary. We establish the theoretical bounds that determine this machine learning algorithm's performance in handling demand shocks. Computational experiments show that this algorithm outperforms the traditional approaches in a variety of situations including large and frequent shocks of the demand mean. The method can also be used as a meta‐algorithm by incorporating other traditional approaches as experts. Working together, the original algorithm and the extended meta‐algorithm can help manufacturers and retailers better adapt their production and inventory control decisions in dynamic environments where demand information is limited and demand shocks are frequent
- 23 September 2015
- Published by Wiley in Decision Sciences
- Vol. 47 (1) , 125-156
Newsvendor Problems with Demand Shocks and Unknown Demand Distributions
In today's competitive market, demand volume and even the underlying demand distribution can change quickly for a newsvendor seller. We refer to sudden changes in demand distribution as demand shocks. When a newsvendor seller has limited demand distribution information and also experiences underlying demand shocks, the majority of existing methods for newsvendor problems may not work well since they either require demand distribution information or assume stationary demand distribution. We present a new, robust, and effective machine learning algorithm for newsvendor problems with demand shocks but without any demand distribution information. The algorithm needs only an approximate estimate of the lower and upper bounds of demand range; no other knowledge such as demand mean, variance, or distribution type is necessary. We establish the theoretical bounds that determine this machine learning algorithm's performance in handling demand shocks. Computational experiments show that this algorithm outperforms the traditional approaches in a variety of situations including large and frequent shocks of the demand mean. The method can also be used as a meta‐algorithm by incorporating other traditional approaches as experts. Working together, the original algorithm and the extended meta‐algorithm can help manufacturers and retailers better adapt their production and inventory control decisions in dynamic environments where demand information is limited and demand shocks are frequent
- Top Cited Papers
- 11 March 2012
- Published by SAGE Publications in Production and Operations Management
- Vol. 21 (6) , 997-1014
Competition and Coordination in Online Marketplaces
Online marketplaces, such as those operated by Amazon, have seen rapid growth in recent years. These marketplaces serve as an intermediary, matching buyers with sellers, whereas control of the good is left to the seller. In some cases, e.g., the Amazon marketplace system, the firm that owns and manages the marketplace system will also sell competing products through the marketplace system. This creates a new form of channel conflict, which is a focus of this article. We consider a setting in which a marketplace firm operates an online marketplace through which retailers can sell their products directly to consumers. We consider a single retailer, who currently sells its product only through its own website, but who may choose to contract with Amazon to sell its product through the marketplace system. Selling the product through the marketplace expands the available market for the retailer, but comes at some expense, e.g., a fixed participation fee or a revenue sharing requirement. Thus, a key question for the retailer is whether she should choose to sell through the marketplace system, and if so, at what price. We analyze the optimal decisions for both the retailer and the marketplace firm and characterize the system equilibrium.
- 11 March 2012
- Published by SAGE Publications in Production and Operations Management
- Vol. 21 (6) , 997-1014
Top Cited PapersCompetition and Coordination in Online Marketplaces
Online marketplaces, such as those operated by Amazon, have seen rapid growth in recent years. These marketplaces serve as an intermediary, matching buyers with sellers, whereas control of the good is left to the seller. In some cases, e.g., the Amazon marketplace system, the firm that owns and manages the marketplace system will also sell competing products through the marketplace system. This creates a new form of channel conflict, which is a focus of this article. We consider a setting in which a marketplace firm operates an online marketplace through which retailers can sell their products directly to consumers. We consider a single retailer, who currently sells its product only through its own website, but who may choose to contract with Amazon to sell its product through the marketplace system. Selling the product through the marketplace expands the available market for the retailer, but comes at some expense, e.g., a fixed participation fee or a revenue sharing requirement. Thus, a key question for the retailer is whether she should choose to sell through the marketplace system, and if so, at what price. We analyze the optimal decisions for both the retailer and the marketplace firm and characterize the system equilibrium.