Wenchang Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Wenchang Zhang has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
Fei Gao
3 shared publicationsLiu Ming
3 shared publications- Yu Guo
International Institute of Finance, School of Management, University of Science and Technology of China, Hefei 230026, China
3 shared publications - Ken Moon
The Wharton School, University of Pennsylvania, Philadelphia, Pennsylvania 19104
2 shared publications - Kostas Bimpikis
Graduate School of Business, Stanford University, Stanford, California 94305
2 shared publications - Liu Ming
School of Management and Economics, The Chinese University of Hong Kong, Shenzhen, Guangdong 518172, China
2 shared publications
Latest Publications
- Open Access
- 13 November 2026
- Published by Elsevier in SSRN Electronic Journal
Delegation at the Doorstep: Courier Negotiation and Locker Pricing in On-Demand Delivery
Food lockers ease the last-100-meter bottleneck in on-demand delivery: couriers avoid doorstep service, while customers collect their meals. Locker suitability depends on two factors unobservable to the platform: customers' pickup hassle and the doorstep friction couriers observe only upon arrival. We compare the Customer Choice Model (CCM), where the fulfillment mode is fixed at checkout, with the Negotiated Fulfillment Model (NFM), where couriers may propose switches from door to locker delivery on arrival, subject to customer consent. Using a stylized model, we characterize the platform's optimal delivery price, courier wage, and, under NFM, locker usage price. We then structurally estimate the model with weighted alternating nonlinear least squares on 101,178 orders from our partner platform. We show that NFM is more profitable than CCM when customers are sufficiently sensitive to waiting: shorter waits from switching high friction orders to lockers support a price increase that outweighs negotiation costs. Contrary to common practice, the platform optimally subsidizes rather than charges couriers for switches when customers are reluctant to pick up from lockers and couriers are costly to recruit. NFM allows the platform to capture value from doorstep friction information it never observes. Granting couriers the right to negotiate raises platform profit and customer surplus but lowers courier welfare, because lighter workloads reduce demand for courier labor. Counterfactuals show that free locker access for couriers, a policy intended to protect them, backfires: relative to CCM, courier welfare falls by 9.66% with free access versus 6.44% with paid access.
- 13 November 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessDelegation at the Doorstep: Courier Negotiation and Locker Pricing in On-Demand Delivery
Food lockers ease the last-100-meter bottleneck in on-demand delivery: couriers avoid doorstep service, while customers collect their meals. Locker suitability depends on two factors unobservable to the platform: customers' pickup hassle and the doorstep friction couriers observe only upon arrival. We compare the Customer Choice Model (CCM), where the fulfillment mode is fixed at checkout, with the Negotiated Fulfillment Model (NFM), where couriers may propose switches from door to locker delivery on arrival, subject to customer consent. Using a stylized model, we characterize the platform's optimal delivery price, courier wage, and, under NFM, locker usage price. We then structurally estimate the model with weighted alternating nonlinear least squares on 101,178 orders from our partner platform. We show that NFM is more profitable than CCM when customers are sufficiently sensitive to waiting: shorter waits from switching high friction orders to lockers support a price increase that outweighs negotiation costs. Contrary to common practice, the platform optimally subsidizes rather than charges couriers for switches when customers are reluctant to pick up from lockers and couriers are costly to recruit. NFM allows the platform to capture value from doorstep friction information it never observes. Granting couriers the right to negotiate raises platform profit and customer surplus but lowers courier welfare, because lighter workloads reduce demand for courier labor. Counterfactuals show that free locker access for couriers, a policy intended to protect them, backfires: relative to CCM, courier welfare falls by 9.66% with free access versus 6.44% with paid access.
- Open Access
- 13 November 2026
- Published by Elsevier in SSRN Electronic Journal
Delegation at the Doorstep: Courier Negotiation and Locker Pricing in On-Demand Delivery
Food lockers ease the last-100-meter bottleneck in on-demand delivery: couriers avoid doorstep service, while customers collect their meals. Locker suitability depends on two factors unobservable to the platform: customers' pickup hassle and the doorstep friction couriers observe only upon arrival. We compare the Customer Choice Model (CCM), where the fulfillment mode is fixed at checkout, with the Negotiated Fulfillment Model (NFM), where couriers may propose switches from door to locker delivery on arrival, subject to customer consent. Using a stylized model, we characterize the platform's optimal delivery price, courier wage, and, under NFM, locker usage price. We then structurally estimate the model with weighted alternating nonlinear least squares on 101,178 orders from our partner platform. We show that NFM is more profitable than CCM when customers are sufficiently sensitive to waiting: shorter waits from switching high friction orders to lockers support a price increase that outweighs negotiation costs. Contrary to common practice, the platform optimally subsidizes rather than charges couriers for switches when customers are reluctant to pick up from lockers and couriers are costly to recruit. NFM allows the platform to capture value from doorstep friction information it never observes. Granting couriers the right to negotiate raises platform profit and customer surplus but lowers courier welfare, because lighter workloads reduce demand for courier labor. Counterfactuals show that free locker access for couriers, a policy intended to protect them, backfires: relative to CCM, courier welfare falls by 9.66% with free access versus 6.44% with paid access.
- 13 November 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessDelegation at the Doorstep: Courier Negotiation and Locker Pricing in On-Demand Delivery
Food lockers ease the last-100-meter bottleneck in on-demand delivery: couriers avoid doorstep service, while customers collect their meals. Locker suitability depends on two factors unobservable to the platform: customers' pickup hassle and the doorstep friction couriers observe only upon arrival. We compare the Customer Choice Model (CCM), where the fulfillment mode is fixed at checkout, with the Negotiated Fulfillment Model (NFM), where couriers may propose switches from door to locker delivery on arrival, subject to customer consent. Using a stylized model, we characterize the platform's optimal delivery price, courier wage, and, under NFM, locker usage price. We then structurally estimate the model with weighted alternating nonlinear least squares on 101,178 orders from our partner platform. We show that NFM is more profitable than CCM when customers are sufficiently sensitive to waiting: shorter waits from switching high friction orders to lockers support a price increase that outweighs negotiation costs. Contrary to common practice, the platform optimally subsidizes rather than charges couriers for switches when customers are reluctant to pick up from lockers and couriers are costly to recruit. NFM allows the platform to capture value from doorstep friction information it never observes. Granting couriers the right to negotiate raises platform profit and customer surplus but lowers courier welfare, because lighter workloads reduce demand for courier labor. Counterfactuals show that free locker access for couriers, a policy intended to protect them, backfires: relative to CCM, courier welfare falls by 9.66% with free access versus 6.44% with paid access.
- 2 November 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
Restaurant Density and Delivery Speed in Food Delivery Platforms
Food delivery platforms (DoorDash, Uber Eats, Meituan) are racing to expand by signing as many restaurants as possible. Such expansion directly translates into changes in restaurant density on the platform, making it crucial to understand how density affects restaurant outcomes. This paper examines the impact of restaurant density on incumbent restaurants and compares three key externalities induced by increased density: customer expansion and cannibalization, which are well documented in the literature, and delivery pooling (i.e., drivers being closer to restaurants and bundling multiple orders before delivery), which has received much less attention. Using a queueing-theoretical model, we develop testable hypotheses on how these externalities influence delivery speed, sales, revenues, and customer spending. Employing two-way fixed effects and instrumental variable estimation with data from a major Chinese food delivery platform, we find that higher restaurant density improves delivery speed and sales for partnered restaurants but reduces order size; heterogeneous-effects estimation further reveals that fast-food and high-volume restaurants benefit more from increased density, while made-to-order and thermally sensitive cuisines see weaker gains. Combining the empirical evidence with our theoretical predictions, our evidence is most consistent with delivery pooling playing a more pivotal role than customer expansion or cannibalization in shaping density effects. Furthermore, nonlinear least squares estimation of our theoretical model reveals that driver arrival speed at restaurants rises with density, shortening customers' wait times and inducing them to place more—but smaller—orders. However, this effect follows an inverted-U-shaped pattern, indicating diminishing delivery pooling benefits—and even adverse effects—at very high density. Finally, we extend our theoretical framework to develop a simulation model, estimated using data, which allows us to isolate the effect of each externality and identify restaurant-specific optimal densities through counterfactual simulations. Results show substantial heterogeneity in sales-maximizing densities across restaurants, with most operating below their optima. Our findings also suggest that platforms should prioritize expansion in low- to moderate-density regions to maximize delivery efficiency and restaurant sales, while carefully managing congestion and tailoring support based on restaurant type in denser markets.
- 2 November 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
Restaurant Density and Delivery Speed in Food Delivery Platforms
Food delivery platforms (DoorDash, Uber Eats, Meituan) are racing to expand by signing as many restaurants as possible. Such expansion directly translates into changes in restaurant density on the platform, making it crucial to understand how density affects restaurant outcomes. This paper examines the impact of restaurant density on incumbent restaurants and compares three key externalities induced by increased density: customer expansion and cannibalization, which are well documented in the literature, and delivery pooling (i.e., drivers being closer to restaurants and bundling multiple orders before delivery), which has received much less attention. Using a queueing-theoretical model, we develop testable hypotheses on how these externalities influence delivery speed, sales, revenues, and customer spending. Employing two-way fixed effects and instrumental variable estimation with data from a major Chinese food delivery platform, we find that higher restaurant density improves delivery speed and sales for partnered restaurants but reduces order size; heterogeneous-effects estimation further reveals that fast-food and high-volume restaurants benefit more from increased density, while made-to-order and thermally sensitive cuisines see weaker gains. Combining the empirical evidence with our theoretical predictions, our evidence is most consistent with delivery pooling playing a more pivotal role than customer expansion or cannibalization in shaping density effects. Furthermore, nonlinear least squares estimation of our theoretical model reveals that driver arrival speed at restaurants rises with density, shortening customers' wait times and inducing them to place more—but smaller—orders. However, this effect follows an inverted-U-shaped pattern, indicating diminishing delivery pooling benefits—and even adverse effects—at very high density. Finally, we extend our theoretical framework to develop a simulation model, estimated using data, which allows us to isolate the effect of each externality and identify restaurant-specific optimal densities through counterfactual simulations. Results show substantial heterogeneity in sales-maximizing densities across restaurants, with most operating below their optima. Our findings also suggest that platforms should prioritize expansion in low- to moderate-density regions to maximize delivery efficiency and restaurant sales, while carefully managing congestion and tailoring support based on restaurant type in denser markets.
- 1 November 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Manufacturing & Service Operations Management
- Vol. 28 (6) , 1749-1766
Reducing Traffic Incidents in Meal Deliveries: Penalize the Platform or Its Independent Drivers?
Problem definition: Some governments have toughened traffic penalties for meal-delivery drivers, yet the number of meal delivery–related traffic incidents has not abated. These observations prompted two questions: Under a given penalty scheme, how should a profit-maximizing platform determine the delivery fee to charge its customers and the commission to pay its drivers? How should a welfare-maximizing government determine its penalty scheme to curb traffic incidents and improve the social surplus of all stakeholders? Methodology/results: We represent the system dynamics using a three-stage game-theoretic model, from which our equilibrium analysis yields three key insights. First, penalizing the platform for delivery-related incidents is more effective at reducing risky driving than penalizing drivers. Higher platform penalties discourage unrealistic delivery-time promises, allowing drivers to lower their driving speeds. By contrast, increasing driver penalties does not consistently curb risky driving, because the platform may offset penalties by raising driver commissions, which further incentivizes speeding. Second, lowering driver penalties expands the platform’s service area by encouraging drivers to accept longer-distance or lower-paying orders. Third, penalizing only the platform—rather than the drivers—maximizes total social surplus, provided the penalty is not set excessively high, which could undermine the balance between safety and market coverage. Our findings remain robust even when key modeling assumptions are relaxed separately. We illustrate our findings using data collected from a Chinese meal-delivery platform. Managerial implications: Governments should shift from penalizing drivers to targeting platforms, acknowledging their pivotal role in shaping driver behavior. This policy change would incentivize platforms to promote safe driving practices over speedy deliveries. Funding: The work is supported by the National Natural Science Foundation of China [Grant 72271212]. Supplemental Material: The online appendices are available at https://doi.org/10.1287/msom.2023.0259 .
- 1 November 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Manufacturing & Service Operations Management
- Vol. 28 (6) , 1749-1766
Reducing Traffic Incidents in Meal Deliveries: Penalize the Platform or Its Independent Drivers?
Problem definition: Some governments have toughened traffic penalties for meal-delivery drivers, yet the number of meal delivery–related traffic incidents has not abated. These observations prompted two questions: Under a given penalty scheme, how should a profit-maximizing platform determine the delivery fee to charge its customers and the commission to pay its drivers? How should a welfare-maximizing government determine its penalty scheme to curb traffic incidents and improve the social surplus of all stakeholders? Methodology/results: We represent the system dynamics using a three-stage game-theoretic model, from which our equilibrium analysis yields three key insights. First, penalizing the platform for delivery-related incidents is more effective at reducing risky driving than penalizing drivers. Higher platform penalties discourage unrealistic delivery-time promises, allowing drivers to lower their driving speeds. By contrast, increasing driver penalties does not consistently curb risky driving, because the platform may offset penalties by raising driver commissions, which further incentivizes speeding. Second, lowering driver penalties expands the platform’s service area by encouraging drivers to accept longer-distance or lower-paying orders. Third, penalizing only the platform—rather than the drivers—maximizes total social surplus, provided the penalty is not set excessively high, which could undermine the balance between safety and market coverage. Our findings remain robust even when key modeling assumptions are relaxed separately. We illustrate our findings using data collected from a Chinese meal-delivery platform. Managerial implications: Governments should shift from penalizing drivers to targeting platforms, acknowledging their pivotal role in shaping driver behavior. This policy change would incentivize platforms to promote safe driving practices over speedy deliveries. Funding: The work is supported by the National Natural Science Foundation of China [Grant 72271212]. Supplemental Material: The online appendices are available at https://doi.org/10.1287/msom.2023.0259 .
- Open Access
- 10 October 2026
- Published by Elsevier in SSRN Electronic Journal
Making the Most of Limited Teacher Time: Personalized Intervention in Online Education
Problem definition: Online learning systems record rich data on student activities, yet teachers still face a hard operational challenge: they must decide which students need support, which skill requires attention, and how to use limited teacher time. Our paper addresses this challenge in collaboration with an online education company. Methodology/results: We first define a binary label of intervention need based on sustained mastery: whether a student masters a skill and continues to show evidence of mastery. Using a wide range of features constructed from historical data, we predict the probability of needing intervention. We then formulate a capacity-constrained scheduling problem that allocates limited teacher time based on these predictions and how long students have waited for intervention. We retrospectively simulate how different scheduling policies would have assigned teacher support over time. The results show our models reduce intervention-delay cost significantly compared with commonly-used instructor heuristics. In addition, models combining a student’s own history with cross-student information perform better than models trained only on the focal student. We also find that scheduling policies that combine predicted intervention need with a modest adjustment for waiting time perform best.We conducted a four-week randomized field experiment that provides evidence that our models increase students’ scores while significantly reducing learning time. Managerial implications: This paper offers a practical approach for education providers seeking to scale teacher support. Better prediction and scheduling can make better use of limited teacher time by balancing predictions of intervention need with waiting times. The partner company has adopted the approach and plans to expand its use.
- 10 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMaking the Most of Limited Teacher Time: Personalized Intervention in Online Education
Problem definition: Online learning systems record rich data on student activities, yet teachers still face a hard operational challenge: they must decide which students need support, which skill requires attention, and how to use limited teacher time. Our paper addresses this challenge in collaboration with an online education company. Methodology/results: We first define a binary label of intervention need based on sustained mastery: whether a student masters a skill and continues to show evidence of mastery. Using a wide range of features constructed from historical data, we predict the probability of needing intervention. We then formulate a capacity-constrained scheduling problem that allocates limited teacher time based on these predictions and how long students have waited for intervention. We retrospectively simulate how different scheduling policies would have assigned teacher support over time. The results show our models reduce intervention-delay cost significantly compared with commonly-used instructor heuristics. In addition, models combining a student’s own history with cross-student information perform better than models trained only on the focal student. We also find that scheduling policies that combine predicted intervention need with a modest adjustment for waiting time perform best.We conducted a four-week randomized field experiment that provides evidence that our models increase students’ scores while significantly reducing learning time. Managerial implications: This paper offers a practical approach for education providers seeking to scale teacher support. Better prediction and scheduling can make better use of limited teacher time by balancing predictions of intervention need with waiting times. The partner company has adopted the approach and plans to expand its use.
- 2 September 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Manufacturing & Service Operations Management
Omnichannel Operations in On-Demand Delivery Platform with Buy-Online-and-Pick-up-in-Store
Problem definition: This paper studies an emerging omnichannel on-demand economy, where consumers decide how their orders are delivered through a platform: fulfilled by independent couriers or picked up in-store by themselves. We analyze the platform’s optimal pricing strategies across service channels and its wage-setting decisions for couriers and assess the resulting welfare implications. Methodology/results: We develop a stylized model to study how buy-online-and-pick-up-in-store (BOPS) influences pricing on both the courier and consumer sides. We then estimate model primitives using data from a leading meal-delivery platform in China and use these estimates to quantify the effects of BOPS. Our main results are as follows. First, although it is widely recognized that price is positively related to demand for gig services, this conventional wisdom no longer holds in an omnichannel environment with BOPS, particularly when consumers are less sensitive to delivery congestion. Second, although BOPS shifts the burden of delivery cost from the platform to consumers, the BOPS-channel price need not be lower than the delivery-channel price in a gig-economy setting. Third, although BOPS always improves platform and merchant profits, it enhances consumer surplus only when market demand is sufficiently great while consistently reducing courier welfare. Managerial implications: In an omnichannel environment, pricing strategies proven effective in the traditional gig economy may become suboptimal. Moreover, because the gains from BOPS are unevenly distributed, platform managers and policymakers should look beyond aggregate gains and consider the welfare losses that BOPS adoption imposes on couriers. Funding: Y. Guo gratefully acknowledges financial support from the National Natural Science Foundation of China [Grants 72471218 and 72188101]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2025.0109 .
- 2 September 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Manufacturing & Service Operations Management
Omnichannel Operations in On-Demand Delivery Platform with Buy-Online-and-Pick-up-in-Store
Problem definition: This paper studies an emerging omnichannel on-demand economy, where consumers decide how their orders are delivered through a platform: fulfilled by independent couriers or picked up in-store by themselves. We analyze the platform’s optimal pricing strategies across service channels and its wage-setting decisions for couriers and assess the resulting welfare implications. Methodology/results: We develop a stylized model to study how buy-online-and-pick-up-in-store (BOPS) influences pricing on both the courier and consumer sides. We then estimate model primitives using data from a leading meal-delivery platform in China and use these estimates to quantify the effects of BOPS. Our main results are as follows. First, although it is widely recognized that price is positively related to demand for gig services, this conventional wisdom no longer holds in an omnichannel environment with BOPS, particularly when consumers are less sensitive to delivery congestion. Second, although BOPS shifts the burden of delivery cost from the platform to consumers, the BOPS-channel price need not be lower than the delivery-channel price in a gig-economy setting. Third, although BOPS always improves platform and merchant profits, it enhances consumer surplus only when market demand is sufficiently great while consistently reducing courier welfare. Managerial implications: In an omnichannel environment, pricing strategies proven effective in the traditional gig economy may become suboptimal. Moreover, because the gains from BOPS are unevenly distributed, platform managers and policymakers should look beyond aggregate gains and consider the welfare losses that BOPS adoption imposes on couriers. Funding: Y. Guo gratefully acknowledges financial support from the National Natural Science Foundation of China [Grants 72471218 and 72188101]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2025.0109 .
- Open Access
- 29 July 2026
- Published by Elsevier in SSRN Electronic Journal
Can AI Replace Human Judgment in Marketplace Operations? Evidence from Keyword Auction Campaigns
We study whether AI can replace human judgment in designing keyword-auction promotion campaigns on a major U.S. e-commerce marketplace. Using comprehensive seller-campaign-auction data for all small and midsize sellers active from 2021-2023, we apply a matched staggered difference-indifferences design to estimate the causal impact of adopting Promoted Listing Campaigns (PLCs) on sellers' weekly net revenues. AI-automated PLCs generate positive returns, increasing weekly net revenues by about $59. However, seller-manual PLCs-configured by sellers using AI recommendations as starting inputs-raise weekly net revenues by an additional $232, primarily through higher sales rather than lower spend. The performance gap reflects AI and sellers solving different optimization problems. The AI maximizes keyword-query relevance (precision), while sellers maximize demand coverage by exploiting private knowledge about how customers actually express purchase intent. By tracing the divergence through the full marketplace production processfrom keyword selection through auction entry, impression generation, clicks, and purchases-we show that manual campaigns sacrifice per-keyword relevance and click-through rate but overwhelm these losses through dramatically broader visibility, with 69.4% more keyword-item pairs entering auctions and generating 149,000 more weekly impressions. A formal model shows that the estimated manual premium identifies the value of sellers' private information, and that sellers rationally sort into manual campaigns when coverage is profitable. The manual advantage is concentrated among high-volume sellers and in complex categories (Electronics, Collectibles), suggesting complementarity rather than substitution between AI automation and human judgment.
- 29 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessCan AI Replace Human Judgment in Marketplace Operations? Evidence from Keyword Auction Campaigns
We study whether AI can replace human judgment in designing keyword-auction promotion campaigns on a major U.S. e-commerce marketplace. Using comprehensive seller-campaign-auction data for all small and midsize sellers active from 2021-2023, we apply a matched staggered difference-indifferences design to estimate the causal impact of adopting Promoted Listing Campaigns (PLCs) on sellers' weekly net revenues. AI-automated PLCs generate positive returns, increasing weekly net revenues by about $59. However, seller-manual PLCs-configured by sellers using AI recommendations as starting inputs-raise weekly net revenues by an additional $232, primarily through higher sales rather than lower spend. The performance gap reflects AI and sellers solving different optimization problems. The AI maximizes keyword-query relevance (precision), while sellers maximize demand coverage by exploiting private knowledge about how customers actually express purchase intent. By tracing the divergence through the full marketplace production processfrom keyword selection through auction entry, impression generation, clicks, and purchases-we show that manual campaigns sacrifice per-keyword relevance and click-through rate but overwhelm these losses through dramatically broader visibility, with 69.4% more keyword-item pairs entering auctions and generating 149,000 more weekly impressions. A formal model shows that the estimated manual premium identifies the value of sellers' private information, and that sellers rationally sort into manual campaigns when coverage is profitable. The manual advantage is concentrated among high-volume sellers and in complex categories (Electronics, Collectibles), suggesting complementarity rather than substitution between AI automation and human judgment.
- 1 February 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 72 (2) , 1112-1131
Marketplace Expansion Through Marquee Seller Adoption: Externalities and Reputation Implications
In the race to establish themselves, many early-stage online marketplaces choose to accelerate their growth by adding marquee (established brand name) sellers. We study the implications of marquee seller entry on smaller, unbranded sellers in a marketplace when both unbranded sellers and marquee sellers can vary vertically across reputation (referred to as sellers’ quality). Although recent literature has shown that higher-quality unbranded sellers fare better than their lower-quality peers, we posit that this may not hold for entrants of any quality. To this end, we collaborate with an online business-to-business platform and exploit the entry of two marquee sellers of vastly differing quality. Using a difference-in-difference-in-differences framework, we causally identify the effect. We find that, although higher-quality unbranded seller revenues increase relative to low-quality unbranded sellers when the entrant is of superior quality (consistent with the literature), the effect is reversed when the entrant is of inferior quality. Further, unbranded sellers change their supply quantities such that the platform’s average supply quality shifts in the direction of entrant quality. Using a stylized theoretical model, we identify two mechanisms that drive our findings: (i) new buyers brought in by the entrant disproportionately favor unbranded sellers who are quality neighbors to the entrant and (ii) the unbranded seller’s ability to adjust their supply quantities. Most notably, the choice of marquee sellers, examined through the lens of their externality on unbranded sellers, can foster or undermine the platform’s long-term growth objectives. This paper was accepted by Victor Martinez de Albéniz, operations management. Supplemental Material: The online appendices and data files are available at https://doi.org/10.1287/mnsc.2022.03606 .
- 1 February 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 72 (2) , 1112-1131
Marketplace Expansion Through Marquee Seller Adoption: Externalities and Reputation Implications
In the race to establish themselves, many early-stage online marketplaces choose to accelerate their growth by adding marquee (established brand name) sellers. We study the implications of marquee seller entry on smaller, unbranded sellers in a marketplace when both unbranded sellers and marquee sellers can vary vertically across reputation (referred to as sellers’ quality). Although recent literature has shown that higher-quality unbranded sellers fare better than their lower-quality peers, we posit that this may not hold for entrants of any quality. To this end, we collaborate with an online business-to-business platform and exploit the entry of two marquee sellers of vastly differing quality. Using a difference-in-difference-in-differences framework, we causally identify the effect. We find that, although higher-quality unbranded seller revenues increase relative to low-quality unbranded sellers when the entrant is of superior quality (consistent with the literature), the effect is reversed when the entrant is of inferior quality. Further, unbranded sellers change their supply quantities such that the platform’s average supply quality shifts in the direction of entrant quality. Using a stylized theoretical model, we identify two mechanisms that drive our findings: (i) new buyers brought in by the entrant disproportionately favor unbranded sellers who are quality neighbors to the entrant and (ii) the unbranded seller’s ability to adjust their supply quantities. Most notably, the choice of marquee sellers, examined through the lens of their externality on unbranded sellers, can foster or undermine the platform’s long-term growth objectives. This paper was accepted by Victor Martinez de Albéniz, operations management. Supplemental Material: The online appendices and data files are available at https://doi.org/10.1287/mnsc.2022.03606 .
- 1 July 2024
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 70 (7) , 4533-4547
Information Provision in Two-Sided Platforms: Optimizing for Supply
Although information design has recently received significant attention as a tool for shaping consumers’ purchase behavior, little is known about its use and implications in two-sided marketplaces, where both supply and demand consist of self-interested strategic agents. In this paper, we develop a model of a two-sided platform that facilitates transactions between vertically differentiated suppliers and consumers who differ in their quality preferences. We focus on illustrating the potential benefits of optimal information provision in managing the supply side of the marketplace, including the entry, exit, and pricing decisions of potential suppliers. We show that the revenue-optimal information provision policy often involves delaying the disclosure of information regarding the quality of the available suppliers. When the outside options available to consumers are relatively unattractive, we find that information design can help the platform achieve a more revenue-efficient matching between supply and demand. On the other hand, when consumers have access to attractive outside options, we show that information design can accelerate the discovery of high-quality suppliers, leading to an increase in the platform’s total transaction volume. Moreover, in cases where the platform benefits from commission subsidies to incentivize the entry of new suppliers, information design can be used to achieve the same goal while allowing the platform to earn higher commission revenue. Overall, our numerical experiments suggest that the impact of information design on platform revenue, as well as on consumer surplus, can be substantial. This paper was accepted by Vishal Gaur, operations management. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mnsc.2023.4894 .
- 1 July 2024
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 70 (7) , 4533-4547
Information Provision in Two-Sided Platforms: Optimizing for Supply
Although information design has recently received significant attention as a tool for shaping consumers’ purchase behavior, little is known about its use and implications in two-sided marketplaces, where both supply and demand consist of self-interested strategic agents. In this paper, we develop a model of a two-sided platform that facilitates transactions between vertically differentiated suppliers and consumers who differ in their quality preferences. We focus on illustrating the potential benefits of optimal information provision in managing the supply side of the marketplace, including the entry, exit, and pricing decisions of potential suppliers. We show that the revenue-optimal information provision policy often involves delaying the disclosure of information regarding the quality of the available suppliers. When the outside options available to consumers are relatively unattractive, we find that information design can help the platform achieve a more revenue-efficient matching between supply and demand. On the other hand, when consumers have access to attractive outside options, we show that information design can accelerate the discovery of high-quality suppliers, leading to an increase in the platform’s total transaction volume. Moreover, in cases where the platform benefits from commission subsidies to incentivize the entry of new suppliers, information design can be used to achieve the same goal while allowing the platform to earn higher commission revenue. Overall, our numerical experiments suggest that the impact of information design on platform revenue, as well as on consumer surplus, can be substantial. This paper was accepted by Vishal Gaur, operations management. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mnsc.2023.4894 .
- Open Access
- 1 January 2024
- Published by Elsevier in SSRN Electronic Journal
Omnichannel Operations in On-Demand Delivery Platform with Buy-Online-and-Pick-up-in-Store
On-demand delivery platforms now provide consumers with the option to buy online and pick up in-store (BOPS). We examine such an omnichannel on-demand economy, where consumers decide how their orders are delivered: by independent couriers or by themselves. First, we develop a stylized model to study the impact of BOPS on platform pricing strategies on both the courier and consumer sides. Our findings reveal new insights that differ from the delivery-only setting. After implementing BOPS, (i) the demand for delivery service decreases, yet the platform may increase price in the delivery channel; (ii) the platform "recruits" consumers to fulfill their own orders, but it may charge BOPS consumers an even higher price than for deliveries; (iii) surge pricing may no longer be optimal in response to a demand surge. Furthermore, the model is extended to incorporate a government-imposed wage floor. Although implementing BOPS reduces courier welfare, we show that policymakers may need to lower the wage floor to increase delivery opportunities for couriers. Finally, we estimate the model primitives using data from a leading meal-delivery platform in China and employ the estimated model to predict the impact of BOPS and wage regulations. We find that BOPS enhances platform profit by 12.00%, whereas it reduces labor welfare by 50.95%. Nevertheless, introducing wage floors alongside BOPS can improve labor welfare more effectively (up to +278.44%) compared to without BOPS (up to +220.14%). Moreover, with BOPS, a small wage floor can significantly increase labor welfare while causing minimal harm to the platform's profit.
- 1 January 2024
- Published by Elsevier in SSRN Electronic Journal
Open AccessOmnichannel Operations in On-Demand Delivery Platform with Buy-Online-and-Pick-up-in-Store
On-demand delivery platforms now provide consumers with the option to buy online and pick up in-store (BOPS). We examine such an omnichannel on-demand economy, where consumers decide how their orders are delivered: by independent couriers or by themselves. First, we develop a stylized model to study the impact of BOPS on platform pricing strategies on both the courier and consumer sides. Our findings reveal new insights that differ from the delivery-only setting. After implementing BOPS, (i) the demand for delivery service decreases, yet the platform may increase price in the delivery channel; (ii) the platform "recruits" consumers to fulfill their own orders, but it may charge BOPS consumers an even higher price than for deliveries; (iii) surge pricing may no longer be optimal in response to a demand surge. Furthermore, the model is extended to incorporate a government-imposed wage floor. Although implementing BOPS reduces courier welfare, we show that policymakers may need to lower the wage floor to increase delivery opportunities for couriers. Finally, we estimate the model primitives using data from a leading meal-delivery platform in China and employ the estimated model to predict the impact of BOPS and wage regulations. We find that BOPS enhances platform profit by 12.00%, whereas it reduces labor welfare by 50.95%. Nevertheless, introducing wage floors alongside BOPS can improve labor welfare more effectively (up to +278.44%) compared to without BOPS (up to +220.14%). Moreover, with BOPS, a small wage floor can significantly increase labor welfare while causing minimal harm to the platform's profit.