7월 4주차

Digital Content Platforms (DCPs), such as Netflix and Spotify, rely on subscriptions and advertising as their primary revenue sources. Traditional revenue models, such as subscription-only and ad-only options, either limit user accessibility or compromise user experience. In response, DCPs have started to combine these two models by offering a two-tier menu: a free ad-supported tier for price-sensitive users and a paid ad-free tier for ad-sensitive users. In this paper, we introduce the Generalized Ad-Supported Subscription (GAS) mechanism-a broad class consisting of subscription-fee and ad-intensity combinations that spans the entire spectrum from fully ad-supported to fully subscription-based access. Using the principles of mechanism design, we characterize a revenue-maximizing GAS mechanism and contrast it with ad-only, subscription-only, and the two-tier menu. Further, we estimate our model parameters and empirically validate our theoretical results in the context of Video-on-Demand platforms. Our analysis shows that the optimal GAS mechanism yields a substantially higher revenue than the ad-only and subscription-only mechanisms. More importantly, this additional revenue improvement does not come at the expense of consumer surplus-this shows that the GAS mechanism improves the overall economic pie by better matching user preferences.

Knowledge-sharing platforms such as Wikipedia are widely used by investors, yet much of the content they circulate is secondhand: factually accurate information that has already been disclosed and is later reshared without substantive updates. Under the efficient market hypothesis (EMH), such information should already be reflected in prices. However, qualitative evidence suggests that retail investors may react to it even when they recognize it as old, implying a behavioral mis-reaction. If systematic, such responses can amplify noise trading and misallocate capital. To provide causal evidence on whether recognized secondhand information influences investor behavior when reshared, we conducted a randomized field experiment on Wikipedia. We focused on salient negative secondhand information, operationalized as litigation news. Between June 15 and July 13, 2013, we introduced such content into the Wikipedia pages of selected public firms and compared their outcomes with those of matched control firms whose pages were left unedited. We also included a placebo group whose pages were edited with operations news. The results show that retail investors respond to recognized secondhand litigation news. Relative to controls, treated firms experienced significant increases in Wikipedia pageviews, a 15.7% rise in retail trading volume, and a 6.9 basis point reduction in bid–ask spreads. These effects were stronger when the litigation was more severe or recent, when firms operated in more visible and positive informational environments, and when the underlying litigation signaled governance weaknesses—a pattern consistent with representativeness-based judgment and therefore difficult to reconcile with the EMH. Mediation analysis further indicates that heightened attention is one channel through which secondhand information shapes trading behavior and market outcomes. We contribute to the information systems and behavioral finance literatures by identifying salient negative secondhand information as a distinct, behaviorally potent content category. Our findings carry implications for regulators, firms, and digital platforms, and point to mitigation strategies such as interface-level cues, investor education, and novelty-detection tools. Overall, we underscore the need to reconsider how platform-mediated salient negative secondhand information can shape market outcomes.

This paper studies how the interplay of posting and trading behaviors within a social network influences an individual’s decision to adopt cryptocurrency. Due to the unique features of cryptocurrency as a financial instrument and the lack of standardized information, social media are an important information channel through which individuals can decide whether to trade cryptocurrencies. However, the vast and unregulated nature of social media can often lead to the dissemination of information of questionable credibility. We exploit a unique dataset from a large social trading platform and study how the interplay of opinions and behavior within social networks influences an individual’s decision to adopt cryptocurrency. We document a pronounced social learning effect in cryptocurrency adoption. Individuals are more inclined to adopt cryptocurrency when their peers do, especially when the trading behavior of those peers aligns with their expressed opinions. Using textual features extracted via a large language model, we show that credibility is the key mechanism underlying the differential effects between adopters and nonadopters. When uncertainty or positive returns are high, investors tend to rely more on positive opinions, regardless of the actual trading behavior of others. Finally, we document the heterogeneous effects of demographic characteristics on social learning. Our study advances social learning theory and has important practical implications.

Modularity undergirds every contemporary platform, yet the contestation of its coordinative capacity conflates complements with substitutes, either undermining app evolution or squandering coordination resources. Without theory to guide them, platforms navigate by costly trial-and-error. Drawing on Simon’s theory of the artifact, we disentangle modularity into three facets—decoupling (minimizing app-platform dependencies), interfaces (enabling app-platform integration), and “design rules” (circumscribing app internals)—theorizing that coordination complements the separability-fostering facets while integration-fostering facets substitute for it. Design rules, ubiquitous in platforms yet theoretically overlooked, generate equivocality through prescriptive fluidity that coordination must resolve. We advance modularity theory by introducing design rules as its third facet—Simon’s missing “inner environment”—showing that coordination’s role is facet-specific. For practice, we equip platform owners to calibrate coordination effort to each facet, particularly at agentic artificial intelligence and hybrid platform frontiers.

When platform owners wield substantial market power, the rules they set often impose unfavorable terms on app developers while restricting their ability to push back through conventional market channels. But platform owners remain vulnerable to challenges from beyond the market. Drawing on a 15-year historical analysis of Apple’s iOS ecosystem (2009–2024), this study explains how large complementors such as Spotify and Epic Games used nonmarket strategies—public advocacy, litigation, and coalition-building—to pressure Apple into changing platform rules, including adaptations to the infamous “Apple tax” and anti-steering provisions. We identify three interlinked mechanisms through which these “complementor giants” orchestrate change: They create public opportunities for stakeholders to voice concerns, generate confluence across separate grievances, and render tensions salient. Together, these mechanisms invite legitimizing responses from regulators, media, users, other developers, and even the platform owner itself. Over time, these dynamics build sustained pressure and momentum for rule change. For practitioners, the findings suggest that complementors in concentrated platform markets may need to look beyond market tactics and invest in carefully designed nonmarket campaigns. For policymakers, the study suggests that rather than prescribing platform owner behavior, regulation could focus on strengthening the conditions under which complementors can advocate for themselves.

Many online platforms now use complex recommender systems to decide which products, restaurants, or services people see. These systems can improve matching, but their recommendations are often hard for users and managers to understand. This article introduces MG-GAT, a recommender system framework that uses multiple networks and attribute data while keeping track of the neighbors and features that influence each recommendation. The same evidence is then used to generate explanations, so the reason shown to a user is tied to the model’s internal decision process rather than added afterward. Across Yelp data from Ontario and Pennsylvania, the method performs competitively with strong deep-learning baselines. In a randomized experiment, explanations based on MG-GAT increased users’ trust, persuasiveness, satisfaction, and engagement relative to similarity-based, social, and SHAP-based explanations. For practice, the results show that platforms do not have to choose between predictive performance and accountable explanations. Recommendation teams can design systems that expose the signals behind predictions, audit generated explanations for unsupported claims, and give users clearer reasons for accepting or questioning automated recommendations.

We develop an end-to-end, scalable machine learning framework for designing bundle recommendations in a high-dimensional choice setting. We leverage historical purchases and consideration sets determined from clickstream data to generate dense representations (embeddings) of products. We impose minimal structure on these embeddings and develop heuristics for complementarity and substitutability among products. Subsequently, we use the heuristics to create multiple bundle recommendations for each of 4,500 focal products and test their performance using a field experiment with a large retailer. We use the experimental data to optimize the recommendation design policy with offline policy learning. Our optimized policy is robust across product categories, generalizes well to the retailer’s entire assortment, and provides an expected improvement of 35% ([Formula: see text] per 100 visits) in revenue from bundle recommendations over the baseline policy. This paper was accepted by Hemant Bhargava, information systems. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2022.01864 .