Song Lin
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
Song Lin has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Michelle Y. Lu
China Europe International Business School, Marketing, Shanghai, Shanghai, 201203, China
2 shared publications - T. Tony Ke
Chinese University of Hong Kong, Hong Kong
2 shared publications Catherine Tucker
Professor of Marketing, MIT Sloan School of Management, Massachusetts Institute of Technology
1 shared publication- Da He
Xiamen University, Xiamen, Fujian 361005, China
1 shared publication Duncan Simester
Professor of Marketing, MIT Sloan School of Management, Massachusetts Institute of Technology
1 shared publicationEric Anderson
Professor of Marketing, Kellogg School of Management, Northwestern University
1 shared publication
Latest Publications
- Top Cited Papers
- 15 June 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
Information Design of Online Platforms
We study how an online platform strategically uses information to both guide consumer search and influence sellers’ targeted advertising. Our model unifies personalized recommendations and targeted ads under an information design framework. We illustrate a fundamental tradeoff facing the platform between improving match efficiency and extracting seller surplus by inducing their competition for prominence. The optimal information design may be socially inefficient because it balances the tradeoff by limiting consumer search and mixing the matched product with a long tail of unmatched ones for recommendation. This implies that sponsored targeted advertising on retail platforms may introduce match inefficiency. This paper was accepted by Greg Shaffer, marketing. Funding: T. T. Ke acknowledges financial support from the National Natural Science Foundation of China [Grants 72422003 and 72394395] and the Hong Kong Research Grants Council [Grant 14503122].
- 15 June 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
Top Cited PapersInformation Design of Online Platforms
We study how an online platform strategically uses information to both guide consumer search and influence sellers’ targeted advertising. Our model unifies personalized recommendations and targeted ads under an information design framework. We illustrate a fundamental tradeoff facing the platform between improving match efficiency and extracting seller surplus by inducing their competition for prominence. The optimal information design may be socially inefficient because it balances the tradeoff by limiting consumer search and mixing the matched product with a long tail of unmatched ones for recommendation. This implies that sponsored targeted advertising on retail platforms may introduce match inefficiency. This paper was accepted by Greg Shaffer, marketing. Funding: T. T. Ke acknowledges financial support from the National Natural Science Foundation of China [Grants 72422003 and 72394395] and the Hong Kong Research Grants Council [Grant 14503122].
- 1 May 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
- Vol. 45 (3) , 515-533
Price Caps by Matching Platforms: The Case of Ticket Resales
This study examines ticket resale platforms’ voluntary price caps, analyzing their adoption incentives and effects on prices, sales, profitability, and welfare.
- 1 May 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
- Vol. 45 (3) , 515-533
Price Caps by Matching Platforms: The Case of Ticket Resales
This study examines ticket resale platforms’ voluntary price caps, analyzing their adoption incentives and effects on prices, sales, profitability, and welfare.
- 11 December 2024
- Published by SAGE Publications in Journal of Marketing
Do More Likes Lead to More Clicks? Evidence from a Field Experiment on Social Advertising
One advantage of advertising on social media is the ability to leverage users’ likes to influence the perceptions and responses of others in their network. Through a large-scale field experiment on WeChat, three online lab studies, and a theoretical model, the authors explore whether and how displaying more likes in an ad can effectively lead to more ad likes and clicks. They find that displaying the first like can significantly increase users’ tendencies to both like and click on an ad. However, on average, showing additional likes does not further increase the clicking propensity, although it consistently attracts more likes. The authors further find that displaying more likes increases the click-through rate for lesser-known brands but not for well-known brands, and it has a stronger impact on the like rate for more socially engaged users than for less socially engaged ones. These findings are consistent with the interplay between informational and normative social influences in social advertising. The public visibility of likes makes liking more susceptible to normative social influence than clicking. The coexistence of these two forces can lead to an enhanced conformity effect on liking and a crowding-out effect on clicking. The findings offer novel implications for managing social advertising and designing social media platforms.
- 11 December 2024
- Published by SAGE Publications in Journal of Marketing
Do More Likes Lead to More Clicks? Evidence from a Field Experiment on Social Advertising
One advantage of advertising on social media is the ability to leverage users’ likes to influence the perceptions and responses of others in their network. Through a large-scale field experiment on WeChat, three online lab studies, and a theoretical model, the authors explore whether and how displaying more likes in an ad can effectively lead to more ad likes and clicks. They find that displaying the first like can significantly increase users’ tendencies to both like and click on an ad. However, on average, showing additional likes does not further increase the clicking propensity, although it consistently attracts more likes. The authors further find that displaying more likes increases the click-through rate for lesser-known brands but not for well-known brands, and it has a stronger impact on the like rate for more socially engaged users than for less socially engaged ones. These findings are consistent with the interplay between informational and normative social influences in social advertising. The public visibility of likes makes liking more susceptible to normative social influence than clicking. The coexistence of these two forces can lead to an enhanced conformity effect on liking and a crowding-out effect on clicking. The findings offer novel implications for managing social advertising and designing social media platforms.
- 1 September 2024
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
- Vol. 43 (5) , 1031-1051
Media Formats of Advertising
A theoretical analysis of the role of media format in advertising.
- 1 September 2024
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
- Vol. 43 (5) , 1031-1051
Media Formats of Advertising
A theoretical analysis of the role of media format in advertising.
- Open Access
- 7 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 912
Information Design of Online Platforms
We consider the strategic use of information by an online platform to both guide consumers' search through product recommendations and influence sellers' targeted advertising decision. Drawing on Bayesian persuasion, we posit that the platform can design a public signal that influences the beliefs of both consumers and sellers. Upon observing the signal, a consumer can conduct a sequential search with perfect recall among the sellers. After visiting a seller, the consumer observes the product price and whether the product is a match or not. Sellers set prices and decide how much to bid in an ad auction for each consumer, where the winner is granted a prominent position. The consumer can obtain the price and match information of the seller in the prominent position at no cost, but incur search cost to visit additional sellers.
- 7 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 912
Open AccessInformation Design of Online Platforms
We consider the strategic use of information by an online platform to both guide consumers' search through product recommendations and influence sellers' targeted advertising decision. Drawing on Bayesian persuasion, we posit that the platform can design a public signal that influences the beliefs of both consumers and sellers. Upon observing the signal, a consumer can conduct a sequential search with perfect recall among the sellers. After visiting a seller, the consumer observes the product price and whether the product is a match or not. Sellers set prices and decide how much to bid in an ad auction for each consumer, where the winner is granted a prominent position. The consumer can obtain the price and match information of the seller in the prominent position at no cost, but incur search cost to visit additional sellers.
- Open Access
- 16 December 2020
- Published by Atlantis Press SARL
- p. 849-855
Research on Cosmetic Consuming Behavior of Female College Students Influenced by Weibo Marketing
- 16 December 2020
- Published by Atlantis Press SARL
- p. 849-855
Open AccessResearch on Cosmetic Consuming Behavior of Female College Students Influenced by Weibo Marketing
- 1 March 2020
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
- Vol. 39 (2) , 317-338
Two-Sided Price Discrimination by Media Platforms
A theory of two-sided markets where media platforms can price discriminate among both advertisers and consumers.
- 1 March 2020
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
- Vol. 39 (2) , 317-338
Two-Sided Price Discrimination by Media Platforms
A theory of two-sided markets where media platforms can price discriminate among both advertisers and consumers.
- Open Access
- 1 October 2015
- Published by SAGE Publications in Journal of Marketing Research
- Vol. 52 (5) , 580-592
Harbingers of Failure
The authors identify customers, termed “Harbingers of failure,” who systematically purchase new products that flop. Their early adoption of a new product is a strong signal that a product will fail—the more they buy, the less likely the product will succeed. Firms can identify these customers through past purchases of either new products that failed or existing products that few other customers purchase. The authors discuss how these insights can be readily incorporated into the new product development process. The findings challenge the conventional wisdom that positive customer feedback is always a signal of future success.
- 1 October 2015
- Published by SAGE Publications in Journal of Marketing Research
- Vol. 52 (5) , 580-592
Open AccessHarbingers of Failure
The authors identify customers, termed “Harbingers of failure,” who systematically purchase new products that flop. Their early adoption of a new product is a strong signal that a product will fail—the more they buy, the less likely the product will succeed. Firms can identify these customers through past purchases of either new products that failed or existing products that few other customers purchase. The authors discuss how these insights can be readily incorporated into the new product development process. The findings challenge the conventional wisdom that positive customer feedback is always a signal of future success.
- Open Access
- 1 January 2015
- Published by Elsevier in SSRN Electronic Journal
Add-On Policies Under Vertical Differentiation: Why Do Luxury Hotels Charge for Internet Whereas Economy Hotels Do Not?
This paper examines firms' product policies when they sell an add-on (e.g., Internet service) in addition to a base product (e.g., hotel rooms) under vertical differentiation (e.g., four- vs. three-star hotels). I show that the role of an add-on differs; higher-quality firms prefer to sell it as optional to discriminate consumers, and lower-quality firms trade off discrimination and differentiation, trying to lower the add-on price to lure consumers from higher-quality rivals. In equilibrium, lower-quality firms' policies are more sensitive to the cost-to-value ratio of an add-on. If the ratio is sufficiently small, firms sell it to all consumers, potentially explaining why lower-end hotels are more likely than higher-end ones to offer free Internet service. Contrary to consensus in the literature, optional add-ons can intensify price competition over consumers who trade off a higher-quality base product versus a lower-quality base including an add-on. Hence, higher-quality firms are incentivized to commit to bundling, while lower-quality firms prefer to commit to not selling it. Add-ons can further reduce lower-quality firms' profits if consumers do not observe the prices because holding up consumers ex post encourages them to switch to higher-quality rivals, which then become better off. Therefore, lower-quality firms are incentivized to advertise add-on prices, and higher-quality firms are not.
- 1 January 2015
- Published by Elsevier in SSRN Electronic Journal
Open AccessAdd-On Policies Under Vertical Differentiation: Why Do Luxury Hotels Charge for Internet Whereas Economy Hotels Do Not?
This paper examines firms' product policies when they sell an add-on (e.g., Internet service) in addition to a base product (e.g., hotel rooms) under vertical differentiation (e.g., four- vs. three-star hotels). I show that the role of an add-on differs; higher-quality firms prefer to sell it as optional to discriminate consumers, and lower-quality firms trade off discrimination and differentiation, trying to lower the add-on price to lure consumers from higher-quality rivals. In equilibrium, lower-quality firms' policies are more sensitive to the cost-to-value ratio of an add-on. If the ratio is sufficiently small, firms sell it to all consumers, potentially explaining why lower-end hotels are more likely than higher-end ones to offer free Internet service. Contrary to consensus in the literature, optional add-ons can intensify price competition over consumers who trade off a higher-quality base product versus a lower-quality base including an add-on. Hence, higher-quality firms are incentivized to commit to bundling, while lower-quality firms prefer to commit to not selling it. Add-ons can further reduce lower-quality firms' profits if consumers do not observe the prices because holding up consumers ex post encourages them to switch to higher-quality rivals, which then become better off. Therefore, lower-quality firms are incentivized to advertise add-on prices, and higher-quality firms are not.
- Open Access
- 1 January 2015
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
- Vol. 34 (1) , 1-19
Learning from Experience, Simply
There is substantial academic interest in modeling consumer experiential learning. However, (approximately) optimal solutions to forward-looking experiential learning problems are complex, limiting their behavioral plausibility and empirical feasibility. We propose that consumers use cognitively simple heuristic strategies. We explore one viable heuristic—index strategies—and demonstrate that they are intuitive, tractable, and plausible. Index strategies are much simpler for consumers to use but provide close-to-optimal utility. They also avoid exponential growth in computational complexity, enabling researchers to study learning models in more complex situations. Well-defined index strategies depend on a structural property called indexability. We prove the indexability of a canonical forward-looking experiential learning model in which consumers learn brand quality while facing random utility shocks. Following an index strategy, consumers develop an index for each brand separately and choose the brand with the highest index. Using synthetic data, we demonstrate that an index strategy achieves nearly optimal utility at substantially lower computational costs. Using IRI data for diapers, we find that an index strategy performs as well as an approximately optimal solution and better than myopic learning. We extend the analysis to incorporate risk aversion, other cognitively simple heuristics, heterogeneous foresight, and an alternative specification of brands.
- 1 January 2015
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Marketing Science
- Vol. 34 (1) , 1-19
Open AccessLearning from Experience, Simply
There is substantial academic interest in modeling consumer experiential learning. However, (approximately) optimal solutions to forward-looking experiential learning problems are complex, limiting their behavioral plausibility and empirical feasibility. We propose that consumers use cognitively simple heuristic strategies. We explore one viable heuristic—index strategies—and demonstrate that they are intuitive, tractable, and plausible. Index strategies are much simpler for consumers to use but provide close-to-optimal utility. They also avoid exponential growth in computational complexity, enabling researchers to study learning models in more complex situations. Well-defined index strategies depend on a structural property called indexability. We prove the indexability of a canonical forward-looking experiential learning model in which consumers learn brand quality while facing random utility shocks. Following an index strategy, consumers develop an index for each brand separately and choose the brand with the highest index. Using synthetic data, we demonstrate that an index strategy achieves nearly optimal utility at substantially lower computational costs. Using IRI data for diapers, we find that an index strategy performs as well as an approximately optimal solution and better than myopic learning. We extend the analysis to incorporate risk aversion, other cognitively simple heuristics, heterogeneous foresight, and an alternative specification of brands.