Research

A general conjecture is that social influence is an important driver of product success and a major contributor to widespread adoption. Many studies confirm the existence of individual social effects and their impact on consumer adoption behavior. Yet other studies show that social influence has a marginal, if any, effect on the overall product reach. In this work, we resolve this discrepancy by analyzing the adoption of 3,315 newly released video games (with user numbers ranging from a few to millions) and exploring the games’ proliferation across the social network on Steam, the world’s largest video game platform. We assess the impact of social influence on a product’s reach by estimating users’ adoption preferences and controlling for homophily.

The findings reveal a complex landscape where the magnitude of social influence and its impact on product reach vary widely among products. We identify three distinct product categories based on their diffusion patterns: 1) products with minimal social influence and below-median reach; 2) medium-success products benefiting significantly from social influence; and 3) blockbusters, exceptionally successful products whose success is not reliant on social influence but on alignment with consumer preferences and the distribution of these preferences among the platform’s user population.

A general conjecture is that successful products attain their popularity through the influence of adopters on their peers and product information disseminating over the social network. Indeed, many studies have confirmed the existence of local peer effects and contagion. But others have shown that peer influence has a marginal, if any, effect on cascades of adoptions. In this work, we study this discrepancy by analyzing video games propagating over the social network of gamers on Steam, the world’s largest video game platform. A major identification problem – distinguishing homophily from peer influence – is a challenge in any peer influence study based on observational data. To overcome this, we introduce a novel method, Revealed Preference-based Matching Estimation, that estimates the impact of peer influence on adoption by using an unsupervised machine-learning algorithm to match product adopters to users based solely on the similarity of their past adoption. This procedure is applied to thousands of products and reveals how peer influence changes over their lifecycle, thus allowing us to draw general conclusions about the entire ecosystem. Results show that most products belong to one of two distinct groups, each exhibiting a characteristic temporal pattern of adoption: products that exhibit substantial peer influence; and products that do not, for which adoption is driven by preferences. Considering the reach of products in each group, surprisingly, we found that local peer effects are stronger in less popular products. Even more surprising is the fact that almost all blockbusters (products adopted by millions of users) did not exhibit substantial peer influence at any stage of their lifecycle. These results shed light on the discrepancy between observed local peer effects and the lack of peer influence in large adoption cascades that are characteristic of successful products.

We challenge the conjecture that the successful dissemination of new products depends on the early involvement of highly active and engaged users. By exploring data from Steam, the world's largest video games platform, with millions of users and thousands of products, we find that blockbuster products can be identified based on the low activity levels of the users interacting with them as early as a few days into the monitoring, or by watching the first few hundred adopters. We replicate our results on 3 other large e-commerce platforms. These results suggest that the attention paid by manufacturers and marketers to heavy users may not guarantee the product’s future success. 

One of the basic features of online social networks is the ability to follow other users to consume their content. The underlying assumption is that following a user should provide substantial value, yet, the value gained from following is typically difficult to estimate. To understand how the value gained from following others affects users' search for influence, we look at eToro, one of the world's most popular social trading platforms. On eToro, users can make well-informed decisions regarding whom to follow and whom not to follow, based on their performance – actual financial implications. In this study, we focus on the dynamics and predictors of seeking and terminating influence through the "Copying" tool, which lets users mirror the investment strategies of other users. The tangible financial implications of this tool make the dynamics of resource allocation and value gained from seeking or terminating influence explicit. 

Our findings highlight not only the trait of tendency to be influenced but also the different dynamic patterns of susceptibility to seek and terminate influence from others. While some users become more influenced over time, showcasing social herding, others choose to invest on their own, indicating a potential learning trajectory. Factors influencing these patterns include past experiences, gender, age, trading intent, annual income, and risk-reward expectations. Our findings expand our understanding of the choices of users to follow and unfollow others on social networks.

More and more digital applications adopted a freemium pricing model. These apps offer their customers a single product that has a basic, free-to-use, limited version and a “premium”, subscribed paid version that grants an improved product experience with additional content or features. To transform users to premium, firms have reverted to adopting traditional marketing methods – by offering time-limited free trials of their premium version. Yet, providing a time-limited free trial raises its own challenges. Making the free trial too long would cannibalize the paying option. Making it too short might not be attractive enough for users to try the premium version or to make the conversion decision at the end of it. The effect of the duration of a free trial is especially pertinent when comparing new and existing users as their initial beliefs and uncertainty levels about the premium version’s quality are heterogenous – it might be simpler to convince new users of the benefits of paying for a premium app, and as such, to offer shorter trial duration, than to existing users who are aware of the value of the free version.

In this work, we use a randomized field experiment on a popular creativity application to study the effect of the duration of a free trial on conversion to premium. We find that a longer duration has an overall positive effect on conversion due to two separate effects: (i) an increase in free trial adoption due to signaling and (ii) an increase in conversion at the end of the free trial due to experience. Further, those effects don’t influence all users similarly. New users, who sign up for a free trial at the time of app download, show an increase in adoption but not in conversion. However, experienced users who use the free version first, show an increase in conversion but not in adoption. These findings suggest that applications should create different offers for different types of users.