No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
An Empirical Analysis of Algorithmic Pricing on Amazon Marketplace
Abstract
The rise of e-commerce has unlocked practical applications for algorithmic pricing (also called dynamic pricing algorithms), where sellers set prices using computer algorithms. Travel websites and large, well known e-retailers have already adopted algorithmic pricing strategies, but the tools and techniques are now available to small-scale sellers as well.
While algorithmic pricing can make merchants more competitive, it also creates new challenges. Examples have emerged of cases where competing pieces of algorithmic pricing software interacted in unexpected ways and produced unpredictable prices, as well as cases where algorithms were intentionally designed to implement price fixing. Unfortunately, the public currently lack comprehensive knowledge about the prevalence and behavior of algorithmic pricing algorithms in-the-wild.
In this study, we develop a methodology for detecting algorithmic pricing, and use it empirically to analyze their prevalence and behavior on Amazon Marketplace. We gather four months of data covering all merchants selling any of 1,641 best-seller products. Using this dataset, we are able to uncover the algorithmic pricing strategies adopted by over 500 sellers. We explore the characteristics of these sellers and characterize the impact of these strategies on the dynamics of the marketplace.
... Chen et al. [10] Investigates usage and effect of algorithmic pricing on Amazon Buy Box offer selection. ...
... Similarly, two prior works discussed seller opportunities and effect of different algorithmic systems or policies on Amazon. Chen et al. [10] also studied Amazon's Buy Box algorithm (similar to what is done in the present study). However, their primary goal was to understand the effects of algorithmic pricing and not the question of potential preferential treatment toward Related Sellers. ...
... This percentage is slightly lower for Germany (50%) and India (40%). This observation not only highlights the importance of price in decision making of customers, but also corroborates the findings of prior works [10] where it was shown that price is among the influential factors in algorithmic decision making as well. Delivery option: Although not explicitly mentioned as many times as some of the other features, 81.2% of the Indian respondents and 66% of the respondents from USA stated that they opt for sellers whose offers are fulfilled by Amazon (including the Amazon SMs). ...
E-commerce marketplaces provide business opportunities to millions of sellers worldwide. Some of these sellers have special relationships with the marketplace by virtue of using their subsidiary services (e.g., fulfillment and/or shipping services provided by the marketplace) -- we refer to such sellers collectively as Related Sellers. When multiple sellers offer to sell the same product, the marketplace helps a customer in selecting an offer (by a seller) through (a) a default offer selection algorithm, (b) showing features about each of the offers and the corresponding sellers (price, seller performance metrics, seller's number of ratings etc.), and (c) finally evaluating the sellers along these features. In this paper, we perform an end-to-end investigation into how the above apparatus can nudge customers toward the Related Sellers on Amazon's four different marketplaces in India, USA, Germany and France. We find that given explicit choices, customers' preferred offers and algorithmically selected offers can be significantly different. We highlight that Amazon is adopting different performance metric evaluation policies for different sellers, potentially benefiting Related Sellers. For instance, such policies result in notable discrepancy between the actual performance metric and the presented performance metric of Related Sellers. We further observe that among the seller-centric features visible to customers, sellers' number of ratings influences their decisions the most, yet it may not reflect the true quality of service by the seller, rather reflecting the scale at which the seller operates, thereby implicitly steering customers toward larger Related Sellers. Moreover, when customers are shown the rectified metrics for the different sellers, their preference toward Related Sellers is almost halved. We believe our findings will inform and encourage further deliberation toward more effective governance of such design choices and policies adopted by e-commerce marketplaces.
... Chen et al. [10] Investigates usage and effect of algorithmic pricing on Amazon Buy Box offer selection. ...
... Similarly, two prior works discussed seller opportunities and effect of different algorithmic systems or policies on Amazon. Chen et al. [10] also studied Amazon's Buy Box algorithm (similar to what is done in the present study). However, their primary goal was to understand the effects of algorithmic pricing and not the question of potential preferential treatment toward Related Sellers. ...
... This percentage is slightly lower for Germany (50%) and India (40%). This observation not only highlights the importance of price in decision making of customers, but also corroborates the findings of prior works [10] where it was shown that price is among the influential factors in algorithmic decision making as well. Delivery option: Although not explicitly mentioned as many times as some of the other features, 81.2% of the Indian respondents and 66% of the respondents from USA stated that they opt for sellers whose offers are fulfilled by Amazon (including the Amazon SMs). ...
E-commerce marketplaces provide business opportunities to millions of sellers worldwide. Some of these sellers have special relationships with the marketplace by virtue of using their subsidiary services (e.g., fulfillment and/or shipping services provided by the marketplace) -- we refer to such sellers collectively as Related Sellers. When multiple sellers offer to sell the same product, the marketplace helps a customer in selecting an offer (by a seller) through (a) a default offer selection algorithm, (b) showing features about each of the offers and the corresponding sellers (price, seller performance metrics, seller's number of ratings etc.), and (c) finally evaluating the sellers along these features. In this paper, we perform an end-to-end investigation into how the above apparatus can nudge customers toward the Related Sellers on Amazon's four different marketplaces in India, USA, Germany and France. We find that given explicit choices, customers' preferred offers and algorithmically selected offers can be significantly different. We highlight that Amazon is adopting different performance metric evaluation policies for different sellers, potentially benefiting Related Sellers. For instance, such policies result in notable discrepancy between the actual performance metric and the presented performance metric of Related Sellers. We further observe that among the seller-centric features visible to customers, sellers' number of ratings influences their decisions the most, yet it may not reflect the true quality of service by the seller, rather reflecting the scale at which the seller operates, thereby implicitly steering customers toward larger Related Sellers. Moreover, when customers are shown the rectified metrics for the different sellers, their preference toward Related Sellers is almost halved.
... The foundation of effective order monitoring lies in the ability to capture and process transactional data instantaneously, enabling organizations to respond proactively rather than reactively to emerging patterns and potential issues. Research analyzing algorithmic pricing on major e-commerce marketplaces has revealed that sellers utilizing realtime monitoring systems can adjust their competitive positioning within minutes rather than hours, providing significant advantages in dynamic pricing environments where conditions change rapidly throughout the day [2]. This capability transforms traditional post-hoc analysis into dynamic, actionable intelligence that can directly impact customer satisfaction, inventory management, and operational efficiency across the entire supply chain. ...
... This comprehensive monitoring becomes particularly valuable during peak shopping periods, when transaction volumes can increase by 400-500% compared to average days, straining conventional order processing systems. According to marketplace analysis, sellers who maintain consistent performance metrics during these high-volume periods see substantially higher customer retention rates in subsequent months [2]. The resulting improvements in fulfillment consistency create measurable impacts on repeat purchase behavior, with studies showing that customers who experience seamless order execution are 15-20% more likely to return to the same platform for future purchases [1]. ...
Real-time order monitoring represents a transformative advancement in e-commerce operations, enabling businesses to track, analyze, and optimize the entire fulfillment process as events occur. This technological evolution has shifted the industry from retrospective batch processing to instantaneous event-driven architectures that capture each customer interaction as it happens. Through specialized components including event producers, message brokers, stream processors, and real-time databases, these monitoring systems deliver unprecedented visibility across the order lifecycle. The integration of advanced analytics further enhances these capabilities, enabling predictive insights and automated responses that address issues before they impact customer experience. Despite technical challenges related to scalability, data consistency, and latency management, innovative solutions have emerged to ensure reliable operation at scale. As these systems mature, emerging technologies such as artificial intelligence, Internet of Things, blockchain, and immersive visualization are extending monitoring capabilities beyond passive observation toward active orchestration of the entire customer journey.
... More precisely, it aims at ensuring their privacy, transparency, fairness and compliance with ethical and legal standards [17,34]. This field is very active, in reaction to algorithms becoming increasingly ubiquitous in our daily lives in critical areas such as finance, human resources, healthcare or justice [1,5,10,12,25,26,30,45,48]. ...
... Fairness auditing, particularly in black-box settings, has been extensively explored [2,12,17,43,45]. These approaches rely on analyzing the outputs of the model based on a set of queries designed to assess fairness. ...
The emergence of AI legislation has increased the need to assess the ethical compliance of high-risk AI systems. Traditional auditing methods rely on platforms' application programming interfaces (APIs), where responses to queries are examined through the lens of fairness requirements. However, such approaches put a significant burden on platforms, as they are forced to maintain APIs while ensuring privacy, facing the possibility of data leaks. This lack of proper collaboration between the two parties, in turn, causes a significant challenge to the auditor, who is subject to estimation bias as they are unaware of the data distribution of the platform. To address these two issues, we present P2NIA, a novel auditing scheme that proposes a mutually beneficial collaboration for both the auditor and the platform. Extensive experiments demonstrate P2NIA's effectiveness in addressing both issues. In summary, our work introduces a privacy-preserving and non-iterative audit scheme that enhances fairness assessments using synthetic or local data, avoiding the challenges associated with traditional API-based audits.
... • Optimizing pricing and inventory: Analytics can help companies dynamically adjust pricing based on demand and optimize inventory levels. Amazon's dynamic pricing algorithms, which change prices millions of times per day, exemplify this capability (Chen et al., 2016). ...
... The deployment of AI/ML and data analytics-backed automation (or even partial automation) across an entire enterprise will effectively navigate the risk of each aspect as the metrics collected will rise to also unprecedented levels. As illustrated by Chen et al. (2016), in the easiest of case studies, pricing and inventory management can be automated to accommodate rise and declines in demand for different products and categories of products or related services in an inventory. This effectively means that the interference of human management into the supply chain management and pricing can be made minimal, freeing managers to work towards not performing minimal necessary tasks to keep operations running but devote time and energy to help grow the business. ...
Digital transformation is reshaping organizations across industries, requiring leaders to adapt to evolving technologies and drive strategic change. This study explores the role of digital leadership in successfully navigating digital transformation, highlighting key competencies, challenges, and strategic approaches. Leaders must develop a digital vision, embrace data-driven decision-making, and manage technology integration, including cloud computing and cybersecurity considerations. The research underscores the significance of leveraging data analytics to optimize operations, improve customer experiences, and drive innovation. Additionally, it examines the evolving role of technology leaders, emphasizing the need for a blend of technical acumen and business strategy. Cybersecurity, a critical component of digital transformation, is analyzed with a focus on risk mitigation and organizational resilience. The paper provides insights into best practices for organizations to adapt and thrive in an increasingly digital world. By fostering a culture of continuous learning, strategic foresight, and cross-functional collaboration, organizations can achieve sustainable growth and maintain competitive advantage.
... For each item, Amazon's proprietary algorithms choose a seller (either itself or a 3p) as the featured seller on its Buy Box. More than 80% of a product's sales are attributed to the Buy Box (Chen et al., 2016). Some of the factors that determine a seller's likelihood to win the Buy Box are price and seller reputation (Chen et al., 2016;Á Gómez-Losada and Duch-Brown, 2019). ...
... More than 80% of a product's sales are attributed to the Buy Box (Chen et al., 2016). Some of the factors that determine a seller's likelihood to win the Buy Box are price and seller reputation (Chen et al., 2016;Á Gómez-Losada and Duch-Brown, 2019). Amazon and 3p sellers change prices dynamically to win the Buy Box. ...
Multivariate random forests (or MVRFs) are an extension of tree-based ensembles to examine multivariate responses. MVRF can be particularly helpful where some of the responses exhibit sparse (e.g., zero-inflated) distributions, making borrowing strength from correlated features attractive. Tree-based algorithms select features using variable importance measures (VIMs) that score each covariate based on the strength of dependence of the model on that variable. In this paper, we develop and propose new VIMs for MVRFs. Specifically, we focus on the variable’s ability to achieve split improvement, i.e., the difference in the responses between the left and right nodes obtained after splitting the parent node, for a multivariate response. Our proposed VIMs are an improvement over the default naïve VIM in existing software and allow us to investigate the strength of dependence both globally and on a per-response basis. Our simulation studies show that our proposed VIM recovers the true predictors better than naïve measures. We demonstrate usage of the VIMs for variable selection in two empirical applications; the first is on Amazon Marketplace data to predict Buy Box prices of multiple brands in a category, and the second is on ecology data to predict co-occurrence of multiple, rare bird species. A feature of both data sets is that some outcomes are sparse — exhibiting a substantial proportion of zeros or fixed values. In both cases, the proposed VIMs when used for variable screening give superior predictive accuracy over naïve measures.
... Companies like Uber and Amazon have successfully implemented dynamic pricing models to optimize revenue and improve customer satisfaction. According to Chen, Mislove, and Wilson (2016), Uber uses data analytics to analyze supply and demand patterns, allowing them to adjust prices dynamically to balance these factors. This not only maximizes revenue but also ensures that consumers can access services when they need them most, thereby enhancing the overall customer experience. ...
... In comparing these findings with previous research, it is evident that the current study aligns with and extends existing knowledge in several ways. For instance, the findings on dynamic pricing and personalized marketing are consistent with the studies by Chen, Mislove, and Wilson (2016) and Wedel and Kannan (2016), respectively. These studies provided foundational insights into the application of data science in marketing, and the current research builds on these insights by providing empirical evidence from practical implementations. ...
Purpose: The purpose of this research is to explore and elucidate the impact of data science on the field of marketing and how it has transformed traditional consumer insights. It aims to identify the key roles and applications of data analytics in marketing strategies and understand their significance in enhancing consumer understanding and decision-making processes. Study design/methodology/approach: This study adopts a comprehensive literature review approach to analyze and synthesize existing research studies, empirical evidence, and industry reports related to data science in marketing. Various sources such as academic journals, industry publications, and reputable online databases are examined to provide a holistic overview of the topic. Findings: The findings of this research highlight that data science has revolutionized marketing by leveraging advanced analytics tools, techniques, and machine learning algorithms to process vast amounts of consumer data. It enables marketers to extract valuable insights from large data sets, detect trends, predict consumer behavior, and personalize marketing strategies. Moreover, data science empowers marketers to make data-driven decisions based on accurate and real-time information, leading to better targeting, increased customer engagement, and improved return on investment (ROI). Originality/value: This research contributes to the existing literature on data science in marketing by offering an up-to-date and comprehensive analysis of the subject. It emphasizes the significance of utilizing data science techniques in marketing strategies and showcases the potential of data analytics in improving consumer insights. The originality of this study lies in its synthesis of diverse research sources, providing practitioners and researchers with valuable insights into the transformative power of data science in marketing and its potential in driving marketing success.
... Our ndings contribute to the ndings of Chen et. al. (2016), suggesting that algorithmic sellers use target strategies such as lowest price, Amazon price and second lowest price. Algorithmic pricing allows rms to receive more feedback, and win Buy Box more frequently and likely suggesting higher sales volumes and more revenue (Chen et. al. 2016). Furthermore, Wieting and Sapi (2021, p.30-33) nds ...
... Our ndings contribute to the ndings of Chen et. al. (2016), suggesting that algorithmic sellers use target strategies such as lowest price, Amazon price and second lowest price. Algorithmic pricing allows rms to receive more feedback, and win Buy Box more frequently and likely suggesting higher sales volumes and more revenue (Chen et. al. 2016). Furthermore, Wieting and Sapi (2021, p.30-33) nds that Buy Box price decreases by 9% if the monopolist seller is algorithmic, compared to a traditional seller. In this vein, Brown and MacKay (2021, p.14) states that " rms with faster pricing technology have persistently lower prices for identical products", indicating an advantage for ...
The rapid progress of “Machine Learning” (ML) and “Artificial Intelligence” (AI) is altering the dynamics of competition such as pricing and production, market regulation and common understanding of economies. This paper examines the complex link between algorithmic pricing strategies and contemporary market dynamics, offering insights into their multi-layered implications. Our findings indicate a systematic relationship between algorithmic pricing and competitive dynamics. Price dispersion grows with the possibility of being an algorithmic seller in online sale of milk products, indicating a non-collusive pricing strategy takes place. On the other hand, quality differences and higher production within a market may result in a collusive pricing scheme in online markets where pricing choices are made by machines.
JEL Classification: C51, D43, L41
... This practice is increasingly common in online retail markets. Chen et al. (2016) estimated that by 2015, algorithms were used in setting prices for roughly one-third of the top 1600 products on Amazon. By 2018, the average product price on Amazon reportedly changed every ten minutes and adapts to market conditions. 1 Since then, an industry has developed around automated pricing software. ...
The rise of algorithmic pricing in online retail platforms has attracted significant interest in how autonomous software agents interact under competition. This article explores the potential emergence of algorithmic collusion - supra-competitive pricing outcomes that arise without explicit agreements - as a consequence of repeated interactions between learning agents. Most of the literature focuses on oligopoly pricing environments modeled as repeated Bertrand competitions, where firms use online learning algorithms to adapt prices over time. While experimental research has demonstrated that specific reinforcement learning algorithms can learn to maintain prices above competitive equilibrium levels in simulated environments, theoretical understanding of when and why such outcomes occur remains limited. This work highlights the interdisciplinary nature of this challenge, which connects computer science concepts of online learning with game-theoretical literature on equilibrium learning. We examine implications for the Business & Information Systems Engineering (BISE) community and identify specific research opportunities to address challenges of algorithmic competition in digital marketplaces.
... The analysis of these games is relevant today because pricing and bidding are increasingly being automated via learning agents. Learning agents are used to bid in display ad auctions, but they are also used by automated agents that set prices on online platforms such as Amazon (Chen et al. 2016). Whether we can expect the dynamics of such multi-agent interactions to converge to an equilibrium or exhibit inefficient price cycles or even chaos, is an economically important question. ...
Understanding the convergence landscape of multi-agent learning is a fundamental problem of great practical relevance in many applications of artificial intelligence and machine learning. While it is known that learning dynamics converge to Nash equilibrium in potential games, the behavior of dynamics in many important classes of games that do not admit a potential is poorly understood. To measure how ''close'' a game is to being potential, we consider a distance function, that we call ''potentialness'', and which relies on a strategic decomposition of games introduced by Candogan et al. (2011). We introduce a numerical framework enabling the computation of this metric, which we use to calculate the degree of ''potentialness'' in generic matrix games, as well as (non-generic) games that are important in economic applications, namely auctions and contests. Understanding learning in the latter games has become increasingly important due to the wide-spread automation of bidding and pricing with no-regret learning algorithms. We empirically show that potentialness decreases and concentrates with an increasing number of agents or actions; in addition, potentialness turns out to be a good predictor for the existence of pure Nash equilibria and the convergence of no-regret learning algorithms in matrix games. In particular, we observe that potentialness is very low for complete-information models of the all-pay auction where no pure Nash equilibrium exists, and much higher for Tullock contests, first-, and second-price auctions, explaining the success of learning in the latter. In the incomplete-information version of the all-pay auction, a pure Bayes-Nash equilibrium exists and it can be learned with gradient-based algorithms. Potentialness nicely characterizes these differences to the complete-information version.
... -world manifestation of this phenomenon can be observed in hub-and-spoke scenarios, where multiple firms utilize similar algorithmic pricing tools provided by third-party platforms. For example, many sellers on Amazon and eBay adopt third-party pricing software instead of developing their own(Chen et al. 2016). Likewise, online travel agencies (OTAs) such as Expedia and Booking.com ...
Nowadays, a significant share of the business-to-consumer sector is based on online platforms like Amazon and Alibaba and uses AI for pricing strategies. This has sparked debate on whether pricing algorithms may tacitly collude to set supra-competitive prices without being explicitly designed to do so. Our study addresses these concerns by examining the risk of collusion when Reinforcement Learning (RL) algorithms are used to decide on pricing strategies in competitive markets. Prior research in this field focused on Tabular Q-learning (TQL) and led to opposing views on whether learning-based algorithms can result in supra-competitive prices. Building on this, our work contributes to this ongoing discussion by providing a more nuanced numerical study that goes beyond TQL, additionally capturing off- and on- policy Deep Reinforcement Learning (DRL) algorithms, two distinct families of DRL algorithms that recently gained attention for algorithmic pricing. We study multiple Bertrand oligopoly variants and show that algorithmic collusion depends on the algorithm used. In our experiments, we observed that TQL tends to exhibit higher collusion and price dispersion. Moreover, it suffers from instability and disparity, as agents with higher learning rates consistently achieve higher profits, and it lacks robustness in state representation, with pricing dynamics varying significantly based on information access. In contrast, DRL algorithms, such as PPO and DQN, generally converge to lower prices closer to the Nash equilibrium. Additionally, we show that when pre-trained TQL agents interact with DRL agents, the latter quickly outperforms the former, highlighting the advantages of DRL in pricing competition. Lastly, we find that competition between heterogeneous DRL algorithms, such as PPO and DQN, tends to reduce the likelihood of supra-competitive pricing.
... Thus, the growth prospects of small-scale sellers with limited stock are relatively lower as platform logics privilege large-scale sellers with a huge inventory at their disposal. [13]. ...
We examine WhatsApp-based reselling practices adopted by small garment sellers in Surat, a textile city in India, as a response to the challenges posed by high commission costs, confusing dashboards, and restrictive rules of global e-commerce platforms. Through interviews and observations, we show how sellers use WhatsApp's popularity to collaborate with women resellers and customers, enabling participation in online commerce bypassing e-commerce platforms. Using the lens of translation, we argue that WhatsApp functions as a tool and site of praxis for sellers who translate the complicated, standardized, and expensive processes of e-commerce platforms that are in English into multimodal, idiomatic, collabora-tive reselling practices. These are undertaken in regional languages on WhatsApp with the help of traders and women resellers economically benefiting everyone while delivering a personalized online shopping experience for customers. We discuss the politics of this translation, examining its impact on the design of e-commerce platforms while also shaping the discourse of reselling as an empowering pathway for women. CCS Concepts • Human-centered computing → Empirical studies in collab-orative and social computing.
... Autonomous pricing algorithms are increasingly widespread, used in contexts ranging from pricing competing products on the Amazon marketplace [Chen et al., 2016] to determining residential real estate rents [Bortolotti, 2023]. There is evidence that these algorithms can behave in a manner that suggests they are learning to collude [Harrington, 2018]. ...
Algorithmic agents are used in a variety of competitive decision settings, notably in making pricing decisions in contexts that range from online retail to residential home rentals. Business managers, algorithm designers, legal scholars, and regulators alike are all starting to consider the ramifications of "algorithmic collusion." We study the emergent behavior of multi-armed bandit machine learning algorithms used in situations where agents are competing, but they have no information about the strategic interaction they are engaged in. Using a general-form repeated Prisoner's Dilemma game, agents engage in online learning with no prior model of game structure and no knowledge of competitors' states or actions (e.g., no observation of competing prices). We show that these context-free bandits, with no knowledge of opponents' choices or outcomes, still will consistently learn collusive behavior - what we call "naive collusion." We primarily study this system through an analytical model and examine perturbations to the model through simulations. Our findings have several notable implications for regulators. First, calls to limit algorithms from conditioning on competitors' prices are insufficient to prevent algorithmic collusion. This is a direct result of collusion arising even in the naive setting. Second, symmetry in algorithms can increase collusion potential. This highlights a new, simple mechanism for "hub-and-spoke" algorithmic collusion. A central distributor need not imbue its algorithm with supra-competitive tendencies for apparent collusion to arise; it can simply arise by using certain (common) machine learning algorithms. Finally, we highlight that collusive outcomes depend starkly on the specific algorithm being used, and we highlight market and algorithmic conditions under which it will be unknown a priori whether collusion occurs.
... These works delve into the anti-competitive concerns in the digital marketplace. Whereas, several other studies look into issues like algorithmic pricing (Chen, Mislove, and Wilson 2016), fairness across different modalities , vaccine misinformation (Juneja and Mitra 2021) on Amazon. ...
Interleaving sponsored results (advertisements) amongst organic results on search engine result pages (SERP) has become a common practice across multiple digital platforms. Advertisements have catered to consumer satisfaction and fostered competition in digital public spaces; making them an appealing gateway for businesses to reach their consumers. However, especially in the context of digital marketplaces, due to the competitive nature of the sponsored results with the organic ones, multiple unwanted repercussions have surfaced affecting different stakeholders. From the consumers' perspective the sponsored ads/results may cause degradation of search quality and nudge consumers to potentially irrelevant and costlier products. The sponsored ads may also affect the level playing field of the competition in the marketplaces among sellers. To understand and unravel these potential concerns, we analyse the Amazon digital marketplace in four different countries by simulating 4,800 search operations. Our analyses over SERPs consisting 2M organic and 638K sponsored results show items with poor organic ranks (beyond 100th position) appear as sponsored results even before the top organic results on the first page of Amazon SERP. Moreover, we also observe that in majority of the cases, these top sponsored results are costlier and are of poorer quality than the top organic results. We believe these observations can motivate researchers for further deliberation to bring in more transparency and guard rails in the advertising practices followed in digital marketplaces.
... In addition to trading and auction markets, pricing algorithms have become widespread in regular product markets 85,86 because they either provide recommendations to human pricing managers 87,88 or entirely dictate pricing for some firms 86,89 . Although pricing algorithms can help firms to scale and respond to changes in demand, they may also generate anti-competition. ...
From fake social media accounts and generative artificial intelligence chatbots to trading algorithms and self-driving vehicles, robots, bots and algorithms are proliferating and permeating our communication channels, social interactions, economic transactions and transportation arteries. Networks of multiple interdependent and interacting humans and intelligent machines constitute complex social systems for which the collective outcomes cannot be deduced from either human or machine behaviour alone. Under this paradigm, we review recent research and identify general dynamics and patterns in situations of competition, coordination, cooperation, contagion and collective decision-making, with context-rich examples from high-frequency trading markets, a social media platform, an open collaboration community and a discussion forum. To ensure more robust and resilient human-machine communities, we require a new sociology of humans and machines. Researchers should study these communities using complex system methods; engineers should explicitly design artificial intelligence for human-machine and machine-machine interactions; and regulators should govern the ecological diversity and social co-development of humans and machines.
... While algorithms are reshaping broader market dynamics, they might also lead to more dominant positions of some actors and new forms of collusion (Li et al., 2021). Research has further shown that algorithms can result in more frequent price changes in online marketplaces, creating increasing pressure to order and regulate algorithmic markets (Cavallo, 2018;Chen et al., 2016). ...
Algorithms are central in shaping platform markets, impacting not only calculation processes but also influencing market actors’ relationships and market structures more broadly. As “autonomous” market devices that are executable or operate without the need for human intervention, algorithms continuously organize, prioritize, and rank product profiles, resulting in personalized exchange offers based on data processing. However, different market actors also actively attempt to influence these processes, turning platform markets into what could be described as algorithmic battlefields. Despite the growing significance of market-based algorithms, research on their role in market shaping remains scarce. This paper explores the role of algorithms in shaping platform markets. By drawing together literatures from market studies and critical algorithm studies, four key algorithmic characteristics in “agencing” markets are identified: as market actors, targets of market contestations, boundary objects, and market systems connectors. Based on these roles, the paper formulates a research agenda for studying algorithmic agencing in platform markets.
... However, this also raises ethical considerations and potential regulatory scrutiny. Entrepreneurs must ensure transparency and fairness in their pricing algorithms (Chen et al., 2016) to maintain customer trust and comply with regulations. For industry practitioners, understanding these dynamics is crucial. ...
The application of artificial intelligence in e-commerce has emerged as a crucial element for the growth of various businesses, elevating the customer experience. Thus, to establish a competitive edge in this e-commerce scenario, it is imperative that online merchants stay updated with the latest marketing strategies and trends. This chapter examines the role of AI in e - commerce, explores the aspects resulting in its adoption, and investigates its impact on e-commerce. This study adopted a qualitative research design by reviewing the existing literature and explicates models for artificial intelligence in the e - commerce industry. Through a review of relevant literature and case studies, this paper argues that artificial intelligence can provide business with a competitive advantage by the following.The primary implications of artificial intelligence are that it can help companies to optimize revenue.
... Питанням теорії та практики формування динамічних цін а також використанню для цієї цілі штучного інтелекту присвячені роботи таких вітчизняних та зарубіжних науковців як Окландер І., Нестеров В., Шиш А., Музиченко Т., Bharadiya J., Le Chen, Wilson C., Mislove A. [1][2][3][4][5][6] та інших. Значний вклад в дослідження теми вносять також публікації спеціалістів з інтернет-маркетингу, які займаються практичними аспектами формування конкурентних цін. ...
Умови онлайн-торгівлі, коли покупці можуть порівнювати ціни на певний товар в багатьох інтернет-магазинах, спонукає до використання динамічних технологій ціноутворення, які враховують велику кількість даних, таких як ціни в конкурентів, запаси, сезонні коливання попиту і багато інших. Оптимізація цін для велетенського асортименту товарів крупних інтернет-магазинів, з урахуванням багатьох даних для кожного найменування, надто трудомістка для ручного ціноутворення і потребує залучення засобів штучного інтелекту, які засвоюють закономірності на основі великих масивів даних і здатні адаптуватися до нової інформації, швидко перевіряючи різні гіпотези та приймаючи найкращі рішення. Дослідження щодо одного з лідерів інтернет-рітейлу, маркетплейсу Amazon, свідчить, що алгоритмічні засоби ціноутворення поширюються не лише на власні товари цього торгівельного майданчика, але і на товари сторонніх продавців, що використовують для продаж сайт amazon.com. При наявності одного найменування товару у декількох продавців, Amazon створює можливості для переважного отримання трафіку потенційних покупців продавцям з найбільш конкурентними умовами продажу. Крім того, всі сторонні продавці мають можливості для використання певних засобів автоматизації та аналітики. На відміну від Amazon, лідер українського інтернет-рітейлу, маркеплейс Розетка, стороннім продавцям, що використовують для продажів сайт rozetka.com.ua, подібних можливостей поки що не надає. Виходячи з конкурентних переваг, які надає учасникам інтернет-ринку впровадження алгоритмічного ціноутворення, слід очікувати все ширшого впровадження штучного інтелекту для формування цін в онлайн-торгівлі. В той же час, цей процес може супроводжуватись витісненням з онлайн–ринку більш дрібних його учасників, зважаючи на технічні, фінансові та організаційні труднощі впровадження такого засобу динамічного ціноутворення.
... These works delve into the anti-competitive concerns in the digital marketplace. Whereas, several other studies look into issues like algorithmic pricing (Chen, Mislove, and Wilson 2016), fairness across different modalities , vaccine misinformation (Juneja and Mitra 2021) on Amazon. ...
Interleaving sponsored results (advertisements) amongst organic results on search engine result pages (SERP) has become a common practice across multiple digital platforms. Advertisements have catered to consumer satisfaction and fostered competition in digital public spaces; making them an appealing gateway for businesses to reach their consumers. However, especially in the context of digital marketplaces, due to the competitive nature of the sponsored results with the organic ones, multiple unwanted repercussions have surfaced affecting different stakeholders. From the consumers' perspective the sponsored ads/results may cause degradation of search quality and nudge consumers to potentially irrelevant and costlier products. The sponsored ads may also affect the level playing field of the competition in the marketplaces among sellers. To understand and unravel these potential concerns, we analyse the Amazon digital marketplace in four different countries by simulating 4,800 search operations. Our analyses over SERPs consisting 2M organic and 638K sponsored results show items with poor organic ranks (beyond 100th position) appear as sponsored results even before the top organic results on the first page of Amazon SERP. Moreover, we also observe that in majority of the cases, these top sponsored results are costlier and are of poorer quality than the top organic results. We believe these observations can motivate researchers for further deliberation to bring in more transparency and guard rails in the advertising practices followed in digital marketplaces.
... E-commerce platforms use personalized pricing to offer discounts or special deals to loyal customers or to encourage purchases from potential high-value customers. This strategy relies heavily on data analytics and machine learning algorithms to predict consumer behavior and set optimal prices (Chen, Mislove, & Wilson, 2016). ...
This study investigates the impact of dynamic pricing promotion strategies on consumer repeat purchase behavior in the United States. Employing a mixed-methods approach, we collected and analyzed quantitative survey data from 300 respondents, as well as qualitative insights from 15 in-depth interviews and two focus groups. Our findings indicate that personalized pricing is the most effective strategy, increasing repeat purchase likelihood by 25%, followed by demand-based pricing at 15%, and time-based pricing at 10%. The effectiveness of these strategies varies across different demographic segments, with personalized pricing being particularly effective among high-income and younger consumers. Additionally, consumer perceptions of fairness, transparency, and trust play a critical role in the acceptance and success of dynamic pricing strategies. The study highlights the need for businesses to prioritize these factors to build long-term customer loyalty and maximize profitability. Future research directions include longitudinal studies, experimental designs, and industry-specific analyses to further explore the dynamics of consumer behavior in response to dynamic pricing.
... Moreover, on average, adopters saw their revenues increase by 8.6 percent, even though the prices they set after adopting the pricing algorithm were 5.7 percent lower. Similarly, Chen, Mislove, and Wilson (2016) found that 543 out of 1,641 Amazon merchants of best-selling products likely used algorithmic pricing on Amazon, but it is unclear what algorithm they used. Cohen, Hahn, Hall, Levitt, and Metcalfe (2016) showed that surge pricing on the UberX serviceset by the platform's algorithm rather than Uber drivershelped to match the demand and supply of ride sharing in real time, leading to an overall $6.8 billion gain of consumer surplus in the U.S. ...
... Moreover, on average, adopters saw their revenues increase by 8.6 percent, even though the prices they set after adopting the pricing algorithm were 5.7 percent lower. Similarly, Chen, Mislove, and Wilson (2016) found that 543 out of 1,641 Amazon merchants of best-selling products likely used algorithmic pricing on Amazon, but it is unclear what algorithm they used. Cohen, Hahn, Hall, Levitt, and Metcalfe (2016) showed that surge pricing on the UberX serviceset by the platform's algorithm rather than Uber drivers -helped to match the demand and supply of ride sharing in real time, leading to an overall $6.8 billion gain of consumer surplus in the U.S. ...
Over the past decade, an increasing number of firms have delegated pricing decisions to algorithms in consumer markets such as travel, entertainment, and retail; business markets such as digital advertising; and platform markets such as ride-sharing. This trend, driven primarily by the increased availability of digital data and developments in information technology, has economic and social consequences that are not yet well understood. The aim of this paper is therefore to examine various implications and challenges of algorithmic pricing for consumers, managers, and regulators. We contribute to the literature by defining and classifying algorithmic pricing, understanding managers' perceptions and adding empirical evidence on its use, raising important considerations for the three stakeholders, and finally outlining research priorities in this area.
... The optimal solutions were obtained by solving the optimization problem (15) to obtain the total profit π M of the manufacturer. Also, the online total profit π O was calculated according to Equation (3) and the total profit π P of the pharmacy was obtained from Equation (5). The data used in this optimization problem are presented in the following table, where P stands for product and mu denotes a money unit. ...
The rise of e-commerce has significantly impacted consumer shopping habits, resulting in profit loss for traditional supply chains. In response to intense competition, numerous companies have transitioned their business models to embrace dual-channel configurations, seeking to captivate customers and increase their market share. Nonetheless, research on decentralized dual-channel supply chain configurations is scarce and predominantly concentrates on single-period pricing. This paper addresses this gap by employing Stackelberg’s game theory to investigate the multi-periodic pricing and remanufacturing decisions within a decentralized dual-channel supply chain with reverse logistics, specialized in the manufacturing and sales of pharmaceutical products. Moreover, this work considers that the online channel pays a sales commission to the pharmacy in return for the provided after-sales services, aiming to incorporate the aspect of sharing revenues. A mathematical formulation is proposed in a multi-periodic environment allowing us to simultaneously maximize the total profits of the manufacturer, the pharmacy and the online channel, by optimizing the pricing and remanufacturing strategies. Numerical analyses examine the customer purchasing preference’s effect on the demand and pricing decisions of each channel, the impact of the collection cost on the optimal remanufacturing strategy, and assess the break-even point of the total profits generated in both channels according to the sales commission. This study’s novelty lies in employing Stackelberg’s game theory to develop a mathematical formulation for the multi-periodic pricing and remanufacturing problem within a decentralized dual-channel supply chain, incorporating a sales commission between both distributors.
This chapter explores the integration of Artificial Intelligence (AI) and the Internet of Things (IoT) in retail, collectively referred to as AIoT, and its transformative impact on in-store customer experiences. By combining AI's ability to personalize customer interactions with IoT's capacity to connect and optimize physical environments, retailers can enhance engagement, improve operational efficiency, and create seamless, data-driven shopping experiences. The chapter examines AIoT's role in driving personalized retail experiences, its impact on customer behavior, and the technological innovations shaping future trends. It also discusses challenges and ethical considerations, including data privacy, cybersecurity, and algorithmic fairness. Ultimately, AIoT is positioned to revolutionize retail in the Experience Economy, offering significant opportunities for innovation and competitive advantage. The chapter concludes with recommendations for retailers to leverage AIoT responsibly while fostering customer trust.
The rapid integration of artificial intelligence (AI) into the platform economy presents unprecedented challenges to legal regulation and management innovation. This paper explores the intersection of economic law, competition policies, algorithmic governance, and market fairness to address these challenges. Through theoretical analysis and empirical evidence, we develop a comprehensive framework for regulating AI-driven platforms. The study incorporates advanced mathematical models to assess the impact of AI algorithms on competition and market behavior. Additionally, we propose innovative management strategies to ensure market fairness and transparency. This research aims to bridge the gap between legal theory and practical governance in the era of AI.
This paper explores the evolution of AI-driven pricing strategies in the automotive and financial services sectors, focusing on dynamic and deal-based pricing models that adapt in real time to shifts in consumer behavior, supply chain limitations, and market fluctuations. We examine how advanced machine learning techniques, including deep learning and reinforcement learning, enable predictive and adaptive pricing solutions that drive customer loyalty, revenue optimization, and transparency. Explainable AI also features prominently, offering transparency to consumers and regulators alike.
This paper explores the evolution of AI-driven pricing strategies in the automotive and financial services sectors, focusing on dynamic and deal-based pricing models that adapt in real time to shifts in consumer behavior, supply chain limitations, and market fluctuations. We examine how advanced machine learning techniques, including deep learning and reinforcement learning, enable predictive and adaptive pricing solutions that drive customer loyalty, revenue optimization, and transparency. Explainable AI also features prominently, offering transparency to consumers and regulators alike.
This article investigates the incentive and ability of a platform to limit the extent of competition between the sellers it hosts. Absent contractual restrictions, a platform has an incentive to ensure competition between the sellers. This incentive can change with the introduction of so‐called platform most‐favored nation clauses (PMFN) that require the online sellers not to offer better conditions on other distribution channels. Such clauses can align the interests between sellers and platforms to restrict competition. I illustrate that a platform can stabilize seller collusion to its own benefit. These results offer a novel rationale to treat PMFNs with scrutiny.
The advent of online selling in Capiz has revolutionized the operations of local businesses by offering a platform to exhibit products to the global market, thereby promoting economic development in the province. This research study sought to determine the effectiveness of online selling, the marketing methods employed by sellers, and the buying behavior of consumers. This study used a descriptive correlational research design to gather information. The study included a total of 110 online consumers and 15 online sellers in the province of Capiz. The participants were selected using purposive sampling. A researcher-made survey questionnaire and interview guide were used in this study. The data were analyzed and interpreted using the percentage, mean, and Pearson r. The findings of the study showed that the level of efficacy of online selling as a whole and in terms of accessibility, convenience, price, product selection, and customer reviews was very effective. Moreover, the level of the sellers’ marketing strategies was very effective. However, the overall level of consumers’ buying behavior was moderately good. Consequently, when categorized according to socio-demographic profile namely, age, sex, civil status, highest educational attainment, and monthly income, consumers’ buying behavior was average. Lastly, a low relationship but significant was observed between the efficacy of online selling and consumers' buying behavior.
The aim of this chapter is to discuss the benefits of AI systems that foster fundamental business transformation. The effects emerging from the literature audit and substantiated in the model testing hereby demonstrate the power of AI systems to equip companies with the tools needed to manage their relationships with customers in an economically feasible manner. In the literature brought to the table, we see a profound discussion on the expected development of AI systems, and the possibility of replacing humans in the near future. Large language model (LLM) incorporating machine learning (ML), deep learning (DL), and natural language processing (NLP) techniques can aid in training on how to collect and handle large amounts of data. Managing such data quickly, correctly, and securely, could generate market intelligence to boost investments and revenues, which any company wants to achieve. AI may encourage more accurate, distinctive, and scalable marketing, personalised businesses (plans) tailored to specific user demands. Having access to a vast array of customer data, AI can refine browsing history and generate personalised strategies for better targeting, resonating with individual(s) demands. In such a way, it is possible to differentiate not only between customers, but also between companies, by offering a unique selling point, USP. Elevating chatbot capacity with the characteristics as emerging hereby to be crucial in AI efficiency and agency, is a prerequisite for delivering the desired USP, to create experience that brings customers to a journey beyond the traditional market space. However, there are some challenges, such as the risk, privacy, and ethics, that need further attention. Moreover, the bias, believability, and authenticity of information exchange invite further exploration. Although, recently developed and implemented AI systems may generate high volume and diverse content. It turns out this content could be fabricated and not necessarily reflect real facts and data. This is a serious issue worth explanation, given the impact it may have on scaling up business and shaping everyday life, as discussed in detail below.
This chapter outlines the outstanding importance of the price system and briefly discusses the corresponding conditions. It examines the extent to which digital technology makes a difference. In a first step, it focuses on how prices are formed in concrete terms and to what extent digitalisation has a transformational effect in this context, e.g. with regard to the new level of transparency, prices set by algorithms, price discrimination and dynamic pricing. The chapter then discusses the consequences for economic policy and explains the extent to which there is a crucial link between the price system and the normative foundations of the economy.
After years of speculation, price discrimination in e-commerce driven by the
personal information that users leave (involuntarily) online, has started
attracting the attention of privacy researchers, regulators, and the press. In
our previous work we demonstrated instances of products whose prices varied
online depending on the location and the characteristics of perspective online
buyers. In an effort to scale up our study we have turned to crowd-sourcing.
Using a browser extension we have collected the prices obtained by an initial
set of 340 test users as they surf the web for products of their interest. This
initial dataset has permitted us to identify a set of online stores where price
variation is more pronounced. We have focused on this subset, and performed a
systematic crawl of their products and logged the prices obtained from
different vantage points and browser configurations. By analyzing this dataset
we see that there exist several retailers that return prices for the same
product that vary by 10%-30% whereas there also exist isolated cases that may
vary up to a multiplicative factor, e.g., x2. To the best of our efforts we
could not attribute the observed price gaps to currency, shipping, or taxation
differences.
Price discrimination, setting the price of a given product for each customer individually according to his valuation for it, can benefit from extensive information collected online on the customers and thus contribute to the profitability of e-commerce services. Another way to discriminate among customers with different willingness to pay is to steer them towards different sets of products when they search within a product category (i.e., search discrimination). Our main contribution in this paper is to empirically demonstrate the existence of signs of both price and search discrimination on the Internet, and to uncover the information vectors used to facilitate them. Supported by our findings, we outline the design of a large-scale, distributed watchdog system that al-lows users to detect discriminatory practices.
Consumers' purchase decisions are increasingly influenced by user-generated online reviews. Accordingly, there has been growing concern about the potential for posting deceptive opinion spam---fictitious reviews that have been deliberately written to sound authentic, to deceive the reader. But while this practice has received considerable public attention and concern, relatively little is known about the actual prevalence, or rate, of deception in online review communities, and less still about the factors that influence it.
We propose a generative model of deception which, in conjunction with a deception classifier, we use to explore the prevalence of deception in six popular online review communities: Expedia, Hotels.com, Orbitz, Priceline, TripAdvisor, and Yelp. We additionally propose a theoretical model of online reviews based on economic signaling theory, in which consumer reviews diminish the inherent information asymmetry between consumers and producers, by acting as a signal to a product's true, unknown quality. We find that deceptive opinion spam is a growing problem overall, but with different growth rates across communities. These rates, we argue, are driven by the different signaling costs associated with deception for each review community, e.g., posting requirements. When measures are taken to increase signaling cost, e.g., filtering reviews written by first-time reviewers, deception prevalence is effectively reduced.
In online markets, a store's reputation is closely tied to its profitability. Sellers' desire to quickly achieve high reputation has fueled a profitable underground business, which operates as a specialized crowdsourcing marketplace and accumulates wealth by allowing online sellers to harness human laborers to conduct fake transactions for improving their stores' reputations. We term such an underground market a seller-reputation-escalation (SRE) market. In this paper, we investigate the impact of the SRE service on reputation escalation by performing in-depth measurements of the prevalence of the SRE service, the business model and market size of SRE markets, and the characteristics of sellers and offered laborers. To this end, we have infiltrated five SRE markets and studied their operations using daily data collection over a continuous period of two months. We identified more than 11,000 online sellers posting at least 219,165 fake-purchase tasks on the five SRE markets. These transactions earned at least 3,452,530. Our study demonstrates that online sellers using SRE service can increase their stores' reputations at least 10 times faster than legitimate ones while only 2.2% of them were detected and penalized. Even worse, we found a newly launched service that can, within a single day, boost a seller's reputation by such a degree that would require a legitimate seller at least a year to accomplish. Finally, armed with our analysis of the operational characteristics of the underground economy, we offer some insights into potential mitigation strategies.
Recently, Uber has emerged as a leader in the "sharing economy". Uber is a "ride sharing" service that matches willing drivers with customers looking for rides. However, unlike other open marketplaces (e.g., AirBnB), Uber is a black-box: they do not provide data about supply or demand, and prices are set dynamically by an opaque "surge pricing" algorithm. The lack of transparency has led to concerns about whether Uber artificially manipulate prices, and whether dynamic prices are fair to customers and drivers. In order to understand the impact of surge pricing on passengers and drivers, we present the first in-depth investigation of Uber. We gathered four weeks of data from Uber by emulating 43 copies of the Uber smartphone app and distributing them throughout downtown San Francisco (SF) and midtown Manhattan. Using our dataset, we are able to characterize the dynamics of Uber in SF and Manhattan, as well as identify key implementation details of Uber's surge price algorithm. Our observations about Uber's surge price algorithm raise important questions about the fairness and transparency of this system.
We describe a new form of online tracking: explicit, yet unnecessary leakage of personal information and detailed shopping habits from online merchants to payment providers. In contrast to Web tracking, online shops make it impossible for their customers to avoid this proliferation of their data. We record and analyse leakage patterns for N = 881 US Web shops sampled from Web users’ actual online purchase sessions. More than half of the sites shared product names and details with PayPal, allowing the payment provider to build up comprehensive consumption profiles across the sites consumers buy from, subscribe to, or donate to. In addition, PayPal forwards customers’ shopping details to Omniture, a third-party data aggregator with an even larger tracking reach. Leakage to PayPal is commonplace across product categories and includes details of medication or sex toys. We provide recommendations for merchants.
eBay is an online marketplace which allows people to easily engage in commerce with one another. Since the market’s online nature precludes many physical cues of trust, eBay has instituted a reputation system through which users accumulate ratings based on their transactions. However, the eBay Feedback System as currently implemented has serious privacy flaws. When sellers leave feedback, buyers’ purchase histories are exposed through no action of their own. In this paper, we describe and execute a series of attacks, leveraging the feedback system to reveal users’ potentially sensitive purchases. As a demonstration, we collect and identify users who have bought gun-related items and sensitive medical tests. We contrast this information leakage with eBay users’ privacy expectations as measured by an online survey. Finally, we make recommendations towards better privacy in the eBay feedback system.
Recent work has developed supervised methods for detecting deceptive opinion spam-fake reviews written to sound authentic and deliberately mislead readers. And whereas past work has focused on identifying individual fake reviews, this paper aims to identify offerings (e.g., hotels) that contain fake reviews. We introduce a semi-supervised manifold ranking algorithm for this task, which relies on a small set of labeled individual reviews for training. Then, in the absence of gold standard labels (at an offering level), we introduce a novel evaluation procedure that ranks artificial instances of real offerings, where each artificial offering contains a known number of injected deceptive reviews. Experiments on a novel dataset of hotel reviews show that the proposed method outperforms state-of-art learning baselines.
Today, many e-commerce websites personalize their content, including Netflix (movie recommendations), Amazon (product suggestions), and Yelp (business reviews). In many cases, personalization provides advantages for users: for example, when a user searches for an ambiguous query such as ``router,'' Amazon may be able to suggest the woodworking tool instead of the networking device. However, personalization on e-commerce sites may also be used to the user's disadvantage by manipulating the products shown (price steering) or by customizing the prices of products (price discrimination). Unfortunately, today, we lack the tools and techniques necessary to be able to detect such behavior.
In this paper, we make three contributions towards addressing this problem. First, we develop a methodology for accurately measuring when price steering and discrimination occur and implement it for a variety of e-commerce web sites. While it may seem conceptually simple to detect differences between users' results, accurately attributing these differences to price discrimination and steering requires correctly addressing a number of sources of noise. Second, we use the accounts and cookies of over 300 real-world users to detect price steering and discrimination on 16 popular e-commerce sites. We find evidence for some form of personalization on nine of these e-commerce sites. Third, we investigate the effect of user behaviors on personalization. We create fake accounts to simulate different user features including web browser/OS choice, owning an account, and history of purchased or viewed products. Overall, we find numerous instances of price steering and discrimination on a variety of top e-commerce sites.
Online marketplaces often contain information not only about products, but also about the people selling the products. In an effort to facilitate trust, many platforms encourage sellers to provide personal profiles and even to post pictures of themselves. However, these features may also facilitate discrimination based on sellers’ race, gender, age, or other aspects of appearance. In this paper, we test for racial discrimination against landlords in the online rental marketplace Airbnb. Using a new data set combining pictures of all New York City landlords on Airbnb with their rental prices and information about quality of the rentals, we show that non-black hosts charge approximately 12% more than black hosts for the equivalent rental. These effects are robust when controlling for all information visible in the Airbnb marketplace. These findings highlight the prevalence of discrimination in online marketplaces, suggesting an important unintended consequence of a seemingly-routine mechanism for building trust.
In various markets where sellers compete in price, price oscillations are observed rather than convergence to equilibrium. Such fluctuations have been empirically observed in the retail market for gasoline, in airline pricing and in the online sale of consumer goods. Motivated by this, we study a model of price competition in which equilibria rarely exist. We seek to analyze the welfare, despite the nonexistence of equilibria, and present welfare guarantees as a function of the market power of the sellers. We first study best response dynamics in markets with sellers that provide a homogeneous good, and show that except for a modest number of initial rounds, the welfare is guaranteed to be high. We consider two variations: in the first the sellers have full information about the buyer's valuation. Here we show that if there are n items available across all sellers and nmax is the maximum number of items controlled by any given seller, then the ratio of the optimal welfare to the achieved welfare will be at most log n/(n-nmax + 1))+1. As the market power of the largest seller diminishes, the welfare becomes closer to optimal. In the second variation we consider an extended model in which sellers have uncertainty about the buyer's valuation. Here we similarly show that the welfare improves as the market power of the larger seller decreases, yet with a worse ratio of n/(n-nmax + 1). Our welfare bounds in both cases are essentially tight. The exponential gap in welfare between the two variations quantifies the value of accurately learning the buyer's valuation in such settings.
Finally, we show that extending our results to heterogeneous goods in general is not possible. Even for the simple class of k-additive valuations, there exists a setting where the welfare approximates the optimal welfare within any non-zero factor only for O(1/s) fraction of the time, where s is the number of sellers.
Opinionated social media such as product reviews are now widely used by individuals and organizations for their decision making. However, due to the reason of profit or fame, people try to game the system by opinion spamming (e.g., writing fake reviews) to promote or to demote some target products. In recent years, fake review detection has attracted significant attention from both the business and research communities. However, due to the difficulty of human labeling needed for supervised learning and evaluation, the problem remains to be highly challenging. This work proposes a novel angle to the problem by modeling spamicity as latent. An unsupervised model, called Author Spamicity Model (ASM), is proposed. It works in the Bayesian setting, which facilitates modeling spamicity of authors as latent and allows us to exploit various observed behavioral footprints of reviewers. The intuition is that opinion spammers have different behavioral distributions than non-spammers. This creates a distributional divergence between the latent population distributions of two clusters: spammers and non-spammers. Model inference results in learning the population distributions of the two clusters. Several extensions of ASM are also considered leveraging from different priors. Experiments on a real-life Amazon review dataset demonstrate the effectiveness of the proposed models which significantly outperform the state-of-the-art competitors.
We consider a single buyer with a combinatorial preference that would like to purchase related products and services from different vendors,where each vendor supplies exactly one product. We study the general case where subsets of products can be substitutes as well as complementary and analyze the game that is induced on the vendors, where a vendor's strategy is the price that he asks for his product. This model generalizes both Bertrand competition (where vendors are perfect substitutes) and Nash bargaining (where they are perfect complements), and captures a wide variety of scenarios that can appear in complex crowd sourcing or in automatic pricing of related products.
We study the equilibria of such games and show that a pure efficient equilibrium always exists. In the case of submodular buyer preferences we fully characterize the set of pure Nash equilibria, essentially showing uniqueness. For the even more restricted "substitutes" buyer preferences we also prove uniqueness over mixed equilibria. Finally we begin the exploration of natural generalizations of our setting such as when services have costs, when there are multiple buyers or uncertainty about the the buyer's valuation, and when a single vendor supplies multiple products.
Random forests are a combination of tree predictors such that each tree depends on the values of a random vector sampled independently and with the same distribution for all trees in the forest. The generalization error for forests converges a.s. to a limit as the number of trees in the forest becomes large. The generalization error of a forest of tree classifiers depends on the strength of the individual trees in the forest and the correlation between them. Using a random selection of features to split each node yields error rates that compare favorably to Adaboost (Y. Freund & R. Schapire, Machine Learning: Proceedings of the Thirteenth International conference, ***, 148–156), but are more robust with respect to noise. Internal estimates monitor error, strength, and correlation and these are used to show the response to increasing the number of features used in the splitting. Internal estimates are also used to measure variable importance. These ideas are also applicable to regression.
We study scenarios where multiple sellers of a homogeneous good compete on
prices, where each seller can only sell to some subset of the buyers.
Crucially, sellers cannot price-discriminate between buyers. We model the
structure of the competition by a graph (or hyper-graph), with nodes
representing the sellers and edges representing populations of buyers. We study
equilibria in the game between the sellers, prove that they always exist, and
present various structural, quantitative, and computational results about them.
We also analyze the equilibria completely for a few cases. Many questions are
left open.
The Internet is composed of multiple economically-independent service providers that sell bandwidth in their networks so as to maximize their own revenue. Users, on the other hand, route their traffic selfishly to maximize their own utility. How does this selfishness impact the efficiency of operation of the network? To answer this question we consider a two-stage network pricing game where service providers first select prices to charge on their links, and users pick paths to route their traffic. We give tight bounds on the price of anarchy of the game with respect to social value--the total value obtained by all the traffic routed. Unlike recent work on network pricing, in our pricing game users do not face congestion costs; instead service providers must ensure that capacity constraints on their links are satisfied. Our model extends the classic Bertrand game in economics to network settings.
Amazon builds up its european marketplace. Internet Retailer
- M Brohan
Coming soon: Toilet paper priced like airline tickets. The Wall Street Journal
- J And Mattioli
Quarterly e-commerce retail sales, 2nd quarter 2015
- R Liu
- X And Weidenhamer
Purchase details leaked to paypal (short paper)
- S Peetz
- T Gunes
- A And
I know what you're buying: Privacy breaches on ebay
- T And
Amazon seller lists book at $23,698,655.93 -- plus shipping. CNN
- J D Sutter
Coming soon: Toilet paper priced like airline tickets
- J And
Measuring price discrimination and steering on e-commerce web sites
- A Soeller
- G Lazer
- D Mislove
- A And
Quarterly e-commerce retail sales
- R Liu
- X And
Amazon hoax coupled with walmart's price matching leads to ridiculously cheap ps4s. Adweek
- M Castillo
Amazon adds 30 million customers in the past year. Geek Wire
- T Duryee
Amazon is absolutely eviscerating other retailers online new survey shows. Re/code
- J D Ray
Coming soon: Toilet paper priced like airline tickets
- J Angwin
- D Mattioli
The buy box cheat sheet. Feedvisor
- T Benmayer
Amazonfresh delays $299/year prime fresh rollout, tweaks subscription model in california. Geek Wire
- T Bishop
Quarterly e-commerce retail sales
- R Denale
- X Liu
- D Weidenhamer
Amazon buy box: The internet's $80 billion sales button
- K Taft D
Amazon may have up to 80 million high-spending prime members worldwide. GeekWire
- T Duryee
Global e-commerce set to grow 25% in 2015. Internet Retailer
- M Lindner
Amazon buy box: The internet's $80 billion sales button. eWeek
- D K Taft
Target expands online price-match policy to include amazon, wal-mart
- P Ziobro