Lit review

In recent years, social media, streaming services and e-commerce sites have radically changed the way consumers find, choose and consume products, content and information. Underlying this change is the recommendation algorithm, a computational tool that seeks to identify, from the many options available to a user, a small set of products, content or information likely to interest them. Over time, as algorithms have become more complex (incorporating, for example, collaborative, content and deep learning approaches), increasing attention has been paid in academic work across marketing, information systems, psychology and computer science to their impact on consumer decision-making. Much of this earlier work focused on the effectiveness of recommendation systems, on, for example, performance of the algorithm and the accuracy of its predictions. But as more people encountered recommendations in their day-to-day lives, scholars started to ask what impact they might have psychologically and behaviorally. Do recommendation algorithms help consumers by cutting through information overload and search costs? Or do they constrain the choice architecture, reinforcing existing preferences, limiting choice and potentially creating "filter bubbles"? And how might, for example, trust in recommendations, their perceived personalization and the element of serendipity support purchase intentions and satisfaction? Taken together, much of this work has raised questions about whether recommendation systems improve the quality of consumer choice, diversity in their consumption and their ability to make decisions. Some scholars argue they help consumers pick up options they would not otherwise have considered. Others warn that, through algorithmic curation, they risk undermining consumer judgment and encouraging similar consumption across large numbers of people. And there are concerns of manipulation, bias and lack of transparency in these systems. Taken together, these factors have made it difficult to assess the impact of recommendations on consumer agency. In this paper, the review draws on previous academic work on the impact of recommendation algorithms on consumer choice. It considers theoretical perspectives, evidence and methods providing the foundation for that work. The purpose is to highlight areas of agreement and disagreement in the existing evidence, to point to gaps in the knowledge and to suggest possible future research on this increasingly important element of consumer decision-making.

The development of e-commerce platforms, streaming services, and social media has led to radical changes in the way in which customers make purchases and decisions of what to watch, listen to or learn. One thing that makes it possible is a recommendation algorithm. It is an algorithm that predicts customer preferences and selects from all available choices certain items that would be interesting to a particular customer. The improvement in algorithms that are now used in recommendation systems has made this issue very popular among scholars who research marketing, information systems, psychology, and computer science. The research done by scientists on recommendation algorithms can be divided into two periods. First, scholars researched the issue and tried to establish whether such algorithms work well. They paid a lot of attention to the effectiveness of the algorithms used. Later, as algorithms began to have greater impact on the behavior of users, scientists began to analyze the way algorithms affected people and their decisions. Thus, many questions about the influence of algorithms on people were raised. For example, do recommendation algorithms actually increase consumer welfare by making their choices easier or rather restrict choice and autonomy? Apart from this, the influence of such algorithms on trust, personalization, and serendipity was examined. At the moment there is still debate on the impact of these algorithms on the consumer choice quality, consumption variety and decision making ability. On one hand, some researchers say that recommendation algorithms give consumers power by helping them find products that are relevant for them. On the other hand, there is a concern that consumers will be deprived of judgment and their consumption will become more homogeneous because of the algorithms. In addition to this, there are ethical problems of recommendation algorithms such as transparency, manipulation and bias. This review

Recommendation algorithms are the building blocks of e-commerce sites, online streaming services and social media networks. They have revolutionized how consumers search for and decide between an exponentially increasing number of products, media and information. Recommendation algorithms help to manage information overload, guiding us towards our preferred product, music track or online video by categorizing all potential products within narrowly defined guidelines, primarily through collaborative filtering, content-based filtering and deep learning techniques. As these recommendation algorithms have advanced over time, from rule-based to increasingly autonomous systems, they have attracted growing scholarly interest in marketing, information systems, psychology and computer science. Initial studies in this field paid primary attention to the technical performance of recommendation systems, putting forward accuracy as a performance metric. As recommendation algorithms were getting more thoroughly integrated into consumer routines, researchers turned their attention to a number of behavioral and psychological outcomes of recommendation algorithms. In particular, they are wondering whether recommendation algorithms are really helpful to consumers in decreasing search costs and information overload, or whether recommendation algorithms are narrowing the available choice architecture in ways that compromise consumer autonomy, competence, or autonomy, or that simply reinforce existing consumer preferences and alleviate the need to seek out a diverse set of options (i.e., the filter bubble problem). Researchers are also considering how recommendation algorithms influence purchase intention and overall consumer satisfaction by leveraging the role of trust, personalization, and serendipity. The larger field However, despite an increasing literature base, there still remain intense debates about the overall impact of these systems on 1. Consumers' decision-making efficiency, quality and time 2. Consumers' diversity of consumption 3. Consumers' long-term decision-making ability. From a consumer empowerment perspective, some researchers have viewed recommendation systems as a means for consumers to access the digitally mediated universe of a platform (a.k.a. Curation) (Goh, 2022). They believe that recommendation algorithms actually help users access information they would not otherwise have found on their own, democratizing information by making it accessible and available to all. This review brings together findings of previous research on the impact of recommendation algorithms on consumer choice, focusing on main theoretical contributions, empirical insights, and methods used. The review seeks to make sense of and delimit agreement and disagreement, while highlighting gaps and suggesting avenues for future research on this important and growing field.

The explosion of e-commerce, streaming, and social media creates a hyper-competitive marketplace where consumers make countless decisions about content and products. The recommendation algorithm has earned a prominent place in this market by predicting consumer preferences to create individualized product and content suggestions. Researchers in fields including marketing, psychology, computer science, and information systems, have studied how recommendation algorithms shape consumer behavior. Prior research focuses on the shortcomings and improvements of recommendation systems. More recently, psychologists and social scientists have expanded the scope to evaluate recommendation systems based on their impact on consumer choice, especially in relation to welfare and ethics. The research explores if recommendation systems enhance consumer welfare, or if they constrain consumer choice by recommending only items within a filter bubble, thereby reducing the range of choices available to the consumer. Further, researchers have studied consumers’ perception of recommendations in conjunction with their intent to purchase. While more studies are being done, there is still a great deal of controversy concerning the body of work in this area. There are some experts in this field who believe that recommendation systems give consumers more choices. Others believe that if recommendation systems are relied upon too much, consumers will become dependent upon the systems and will stop making their own choices. It is possible that in the future, recommendation systems will be widely used. If this happens, the consequences, both good and bad, will affect consumers globally. In this paper, the effects of recommendation systems on consumer choices will be reviewed. This review will analyze and outline various theories concerning recommendation systems and consumer choices. Additional aspects of this review will include the methods and data collected in various recommendation systems studies. This paper hopes to achieve an understanding of the major viewpoints and minor viewpoints pertaining to recommendation systems. It is the goal of this review to achieve a more complete knowledge of this area and to suggest and prioritize areas for further research in this field.

The spread of e-commerce platforms, streaming services and social media networks has fundamentally changed the way consumers discover and choose products, content and information. In the centre of this change is recommendation algorithm—a computational mechanism to predict user preferences and filter mass of choices in a personalised way into satisfactory options. The evolution of algorithms towards those based on collaborative filtering, content-based filtering and even deep learning combined with their potential influence on consumer decision-making have sparked significant academic interest across a number of fields including marketing, information systems, psychology and computer science. Initial studies in this area mostly concentrated on the technical effectiveness of recommender systems, focusing on accuracy scores and performance at an algorithmic level. But as these systems have become more and more a part of day to day consumer experiences, researchers started examining the potential behavioral and psychological ramifications. With only a few exceptions, most appear pretty good (e.g. we observe the question of whether recommendation algorithms actually serve to benefit consumers by lowering search costs and information overload or instead limit choice architecture in ways that undermine autonomy, induce similar preferences, and make filter bubbles). At the same time, studies have investigated the interplay of algorithmic recommendations with trust, perceived personalization and serendipity in influencing purchase intentions as well as attitudinal satisfaction. Nonetheless, although the literature on CA systems is growing, major controversial discussions are ongoing regarding how their net effects influence quality and diversity of consumer choice as well as long-term skills to make sound decisions. We will explore the benefits and detriments of recommendation. Some scholars argue that recommendation algorithms actually empower consumers by pushing relevant options to them they might not come across, while others warn that dependency on algorithmic curation threatens critical thinking around consumption so we move into echophelia where we all consume the same rubbish as each other. Moreover, transparency, manipulation and algorithmic bias have made it hard to grasp what these systems do with consumer agency. This review integrates previous research on the effects of recommendation algorithms on consumer choice, applying essential theories, evidence from empirical studies, and approaches to methodology. This purportedly serves to clarify the existing areas of agreement and disagreement, discover the missing knowledge in relation to this emergent but burgeoning area of consumer behavior, and suggest new avenues for future research into this increasingly important area of consumer behavior.

The rise of online shopping, streaming services as well as social media has changed the way consumers find and choose products, content, and information. At the core of this process lies the recommendation algorithm, which is a system that predicts user preferences and filters the many possible options that may have been available. As these algorithms got more and more advanced due to collaborative filtering as well as content filtering techniques, interest in the influence recommendation algorithms exert on consumers increased among scholars across a number of different areas. Initially, this research was more focused on the performance of recommendation systems in terms of effectiveness and accuracy, but once these systems were introduced into everyday practice, the research shifted to behavioral and psychological implications of recommendation systems. The questions were raised whether recommendation algorithms are actually increasing consumer welfare in terms of access to the information by offering consumers a wide selection of options available or if they are actually limiting choice architectures resulting in putting consumers in a narrow box in terms of the decision-making. Moreover, researchers looked at algorithmic recommendations in interaction with trust, perceived personalization as well as serendipity in terms of purchase intentions and product satisfaction. Nonetheless, the scholarship is not over as a number of controversies exist in terms of the net effect of recommendation systems on consumer choice behavior as well as consumption diversity. Some scholars believe that recommendation algorithms enhance the decision-making process while others warn that they may lead to the loss of independent judgment and, therefore, homogeneity in terms of consumption. Furthermore, ethical and moral considerations regarding the recommendation systems have appeared in research as well. The article proposes a review of recent studies that investigate the effects of recommendation algorithms on consumer choice and consumption process. approaches. In doing so, it aims to clarify areas of consensus and contention, identify gaps in current knowledge, and propose directions for future research into this increasingly consequential domain of consumer behavior.

The growth of e-commerce and the abundance of social networks and streaming services have changed consumer behavior in many ways. For example, the recommendation algorithm is an invention that has allowed consumers to choose goods and services with more comfort and customization. A scientific field has been devoted to studying the impact of such algorithms on the consumer, and several reviews have analyzed different topics related to recommendation systems. Earlier studies focused mainly on practical applications of recommendation algorithms. However, recently, the topic of the impact of such algorithms on customer behavior has become more popular among academics and companies. Although algorithms generally help customers find exactly what they are looking for, questions remain about the hidden influence of algorithms on the choice of goods and services. Moreover, many works address the question of whether the algorithm is beneficial for the consumer in terms of quality of choice and diversity of choice. While some research highlights the advantages of using recommendation algorithms for both companies and customers, questions arise about the possible negative effects of such algorithms on society and, in particular, on consumer behavior. Moreover, while many works have analyzed the impact of algorithms on the choice of goods and services, the issue of the relationship between algorithms and consumer trust remains a topic of research. This article is a meta-analysis of several works dedicated to the theme of the impact of recommendation algorithms on consumer choice. The purpose of this article is to highlight the main ideas and arguments of the works analyzed and provide a basis for further research on the topic.

With the rapid growth of online platforms, such as e-commerce sites, streaming platforms, and social media platforms, consumers’ product, content, and information choices have significantly changed. Recommendation algorithms are the tools that assist consumers in making choices based on users’ preferences from a large number of items. As these algorithms have become more advanced and complex, there has been much research interest in how recommendation algorithms influence consumer choice. However, there is much discussion on the effects of recommendation algorithms on consumer choice. Early studies on recommendation algorithms have mainly focused on the effectiveness of algorithms in terms of accuracy. However, as recommendation algorithms have been more integrated into consumers’ daily lives, there has been increased interest in how recommendation algorithms affect consumers’ choice. This study aims to review the current state of knowledge regarding how recommendation algorithms affect consumer choice.

The explosive growth of e-commerce portals, streaming services and social networking sites has revolutionized the ways consumers discover and select products, information and services. Central to this transformation is the recommendation algorithm, a computational tool that predicts consumer preferences and narrows the overwhelming variety of available options to a manageable, personalized subset. With increasing sophistication based on collaborative filtering, content-based filtering and methods of machine learning, these algorithms now exert substantial influence on consumer decision processes and have attracted considerable attention from researchers in marketing, information systems, psychology and computer science. Initial work in this area concentrated primarily on the technical performance of recommendation systems and on measures of accuracy and algorithmic efficiency. However, as these systems became more deeply integrated into everyday consumer experience, investigators extended their analyses to include broader behavioral and psychological effects. Questions arose regarding whether recommendation algorithms truly increase consumer welfare by reducing costs of search and information overload, or instead restrict the architecture of choice to the extent of limiting consumer autonomy, reinforcing pre-existing preferences and generating filter bubbles. Finally, the research examines the ways in which algorithmic recommendations modify perceived levels of trust, personalization and serendipity to influence consumer intentions and satisfaction. Although the volume of published work continues to grow, major controversies remain concerning the overall effects of these systems on quality of consumer choice, diversity of consumption experiences and capacity for long-term decision making. Some investigators believe that these algorithms enhance consumer choice by providing access to options that would otherwise escape notice, whereas others contend that excessive reliance on algorithmic modes of information selection impairs capacity for independent judgment and promotes uniformity of patterns of consumption. Also, concerns about the degree of transparency, potential for manipulation and extent of algorithmic bias further impair our understanding of the ways in which these systems influence consumer autonomy. This review integrates and summarizes the major results of previous work on the effects of recommendation algorithms on consumer behavior. The intent is to clarify points of agreement and disagreement, identify major gaps in knowledge and suggest directions for future research on this rapidly expanding area of consumer behavior.

Online shopping sites, streaming services, and social media have completely changed how people find and choose products, content, and information. The main reason for this shift is the recommendation algorithm. These are computer systems that try to guess what users will like and narrow down the huge number of choices to a smaller, personalized list. As these algorithms have gotten more advanced, using methods like collaborative filtering, content-based filtering, and deep learning, their impact on consumer decisions has become a major research topic for marketing, information systems, psychology, and computer science. Early studies mainly looked at how well recommendation systems worked technically, focusing on how accurate they were and how well the algorithms performed. But as these systems became a bigger part of daily life for consumers, researchers started looking into their wider effects on behavior and psychology. Questions have come up about whether recommendation algorithms truly help consumers by making it easier and less overwhelming to find things, or if they actually limit choices by reinforcing current preferences and creating "filter bubbles." At the same time, researchers have studied how algorithmic recommendations affect trust, how personalized they feel, and whether they introduce unexpected discoveries, all of which influence whether people decide to buy something and if they are satisfied. Even though there's a lot of research now, there are still big disagreements about the overall impact of these systems on the quality of consumer choices, the variety of what people consume, and their ability to make decisions over time. Some researchers believe recommendation algorithms help consumers by showing them relevant options they might not have found otherwise. Others worry that relying too much on what algorithms suggest could weaken independent thinking and lead to more similar consumption habits. On top of this, worries about how transparent these algorithms are, whether they can be used to manipulate people, and issues of bias have made it harder to understand how these systems affect consumer freedom. This review brings together current research on how recommendation algorithms affect consumer choice. It looks at the main theories, research findings, and study methods used. By doing this, it hopes to make clear what researchers agree on and where they disagree, point out what we still need to learn, and suggest areas for future research in this growing and important part of consumer behavior.

The proliferation of e-commerce platforms, streaming services, and social media networks has fundamentally transformed how consumers discover and select products, content, and information. At the heart of this transformation lies the recommendation algorithm—a computational system designed to predict user preferences and filter the vast array of available options into a manageable, personalized subset. As these algorithms have become increasingly sophisticated, incorporating collaborative filtering, content-based filtering, and deep learning techniques, their influence over consumer decision-making has drawn substantial scholarly attention across marketing, information systems, psychology, and computer science disciplines. At first, people studying recommendation systems were mainly interested in how well they worked from a technical standpoint, looking at how accurate they were and how good the algorithms were. But as these systems became a bigger part of our daily lives, researchers started to think about how they affect us in a broader sense, like how they impact our behavior and thoughts. Now, people are wondering if these systems really help us by making it easier to find what we want and reducing the amount of information we have to sift through, or if they actually limit our choices and make us stick to what we already like, creating a kind of "filter bubble". At the same time, researchers are also looking at how these algorithmic recommendations interact with how much we trust them, how personalized they seem, and how often they surprise us, and how all these things together influence what we buy and how happy we are with our purchases. There's a lot of talk about how recommendation systems affect the choices we make as consumers. Some people think these systems are helpful because they show us things we might not have found on our own. On the other hand, others are worried that if we rely too much on these algorithms, we'll lose our ability to make decisions for ourselves and end up doing the same things as everyone else. It's also a problem that we don't always know how these systems work or what's really going on behind the scenes. This lack of transparency, along with concerns about manipulation and bias, makes it hard to understand how these systems are really affecting us as consumers. For instance, are they really giving us what we want, or are they just pushing us towards certain things? Do they help us discover new things, or do they just keep showing us the same old stuff? These are important questions because they get at the heart of how much control we have over our own choices. This study looks at how recommendation algorithms affect the choices people make when they buy things. It checks out the main ideas, what research has found, and how researchers do their studies. The goal is to figure out what everyone agrees on and what they don't, find out what we still don't know, and suggest where future research should go. This is a big deal because recommendation algorithms are becoming more and more important in how people make decisions about what to buy.