This paper presents an analysis of a person-to-person recommendation network consisting of 4 million people who made 16 million recommendations on half a million products. The study observes the propagation of recommendations and cascade sizes, which are explained by a simple stochastic model. It analyzes how user behavior varies within user communities defined by the recommendation network. Product purchases follow a 'long tail' where a significant share of purchases belongs to rarely sold items. The study establishes how the recommendation network grows over time and how effective it is from the sender and receiver perspectives. While on average recommendations are not very effective at inducing purchases and do not spread very far, a model is proposed that successfully identifies communities, product, and pricing categories for which viral marketing is very effective. The study finds that product purchases resulting from recommendations are not far from the 80-20 rule, where the top 20% of products account for about half the sales. Effectively advertising niche products using traditional advertising is impractical, so targeted marketing is advantageous for both merchants and consumers. Online product and merchant reviews, as well as collaborative filtering recommendations, help consumers discover new products and receive more accurate evaluations, but cannot completely substitute personalized recommendations from friends or relatives. The study directly observes the effectiveness of person-to-person word of mouth advertising for hundreds of thousands of products for the first time. It finds that most recommendation chains do not grow very large, often terminating with the initial purchase of a product. However, occasionally a product will propagate through a very active recommendation network. A simple stochastic model is proposed to explain the propagation of recommendations. The characteristics of recommendation networks influence the purchase patterns of their members. For example, individuals' likelihood of purchasing a product initially increases as they receive additional recommendations for it, but a saturation point is quickly reached. Interestingly, as more recommendations are sent between the same two individuals, the likelihood that they will be heeded decreases. The study finds that communities, automatically found by graph-theoretic community finding algorithms, were usually centered around a product group, such as books, music, or DVDs, but almost all of them shared recommendations for all types of products. It also finds patterns of homophily, the tendency of like to associate with like, with communities of customers recommending types of products reflecting their common interests. The study proposes models to identify products for which viral marketing is effective. It finds that the category and price of a product play a role, with recommendations of expensive products of interest to small, well-connected communities resulting in a purchase more often. It also observes patterns in the timing of recommendations and purchases corresponding to times of day when people are likely to be shopping online or reading email. The paper discusses the implications of these findings and concludes that the study provides insights into the influence of social networks on purchasing decisions and the effectiveness of viral marketing strategies.This paper presents an analysis of a person-to-person recommendation network consisting of 4 million people who made 16 million recommendations on half a million products. The study observes the propagation of recommendations and cascade sizes, which are explained by a simple stochastic model. It analyzes how user behavior varies within user communities defined by the recommendation network. Product purchases follow a 'long tail' where a significant share of purchases belongs to rarely sold items. The study establishes how the recommendation network grows over time and how effective it is from the sender and receiver perspectives. While on average recommendations are not very effective at inducing purchases and do not spread very far, a model is proposed that successfully identifies communities, product, and pricing categories for which viral marketing is very effective. The study finds that product purchases resulting from recommendations are not far from the 80-20 rule, where the top 20% of products account for about half the sales. Effectively advertising niche products using traditional advertising is impractical, so targeted marketing is advantageous for both merchants and consumers. Online product and merchant reviews, as well as collaborative filtering recommendations, help consumers discover new products and receive more accurate evaluations, but cannot completely substitute personalized recommendations from friends or relatives. The study directly observes the effectiveness of person-to-person word of mouth advertising for hundreds of thousands of products for the first time. It finds that most recommendation chains do not grow very large, often terminating with the initial purchase of a product. However, occasionally a product will propagate through a very active recommendation network. A simple stochastic model is proposed to explain the propagation of recommendations. The characteristics of recommendation networks influence the purchase patterns of their members. For example, individuals' likelihood of purchasing a product initially increases as they receive additional recommendations for it, but a saturation point is quickly reached. Interestingly, as more recommendations are sent between the same two individuals, the likelihood that they will be heeded decreases. The study finds that communities, automatically found by graph-theoretic community finding algorithms, were usually centered around a product group, such as books, music, or DVDs, but almost all of them shared recommendations for all types of products. It also finds patterns of homophily, the tendency of like to associate with like, with communities of customers recommending types of products reflecting their common interests. The study proposes models to identify products for which viral marketing is effective. It finds that the category and price of a product play a role, with recommendations of expensive products of interest to small, well-connected communities resulting in a purchase more often. It also observes patterns in the timing of recommendations and purchases corresponding to times of day when people are likely to be shopping online or reading email. The paper discusses the implications of these findings and concludes that the study provides insights into the influence of social networks on purchasing decisions and the effectiveness of viral marketing strategies.