About three years ago, my colleague Yuping Liu and I were curious about how some YouTube videos gain popularity so quickly while others never get a second look. At this time, videos like "Charlie bit me" and "Chocolate Rain" were rising everyday people to stardom. What made these videos stand out more than others? We set out to answer that question by studying the user-generated content on YouTube.
I'm pleased to announce that today, our research paper, "Rising to Stardom: An Empirical Investigation of the Diffusion of User-generated Content" was published in the Journal of Interactive Marketing. Here is the full-text article.
With the explosive growth of online user-generated content and the desire by marketers to better utilize this space, it is beneficial to understand the viral diffusion of such content and to identify messages that are most likely to achieve popularity. In this paper, we combine network analysis and the diffusion literature to study the spreading of user-generated videos online. We identify three groups of factors that affect diffusion outcomes: network structure, content characteristics, and author characteristics. Using a proportional rates model, we analyze the diffusion of a sample of videos on YouTube. Our results show that it is preferable to have many subscribers who each has a few friends than to have a few subscribers with many connections. Furthermore, a curvilinear relationship exists between subscriber network connectivity and diffusion rate such that diffusion is at its highest under moderate connectivity. Examining content characteristics, we show that entertainment and educational values affect diffusion but production quality does not matter. Moreover, we find that quality as manifested by user ratings influences diffusion more than innate content quality. Not surprisingly, an author's past success carries over to the current content, and content from younger authors is more popular.
We collected a random sample of over 100 user-generated videos newly uploaded onto YouTube over the course of a week. We tracked each video for a period of two months, recording the number of views and the average user ratings each day. We also collected a large number of characteristics for each video, including those related to the video content, to the video author and to the network of users connected to the video author. We had study participants rate each video on its production quality, educational value and entertainment value. Here are some highlights of our findings.
- Diffusion rate is at its highest under moderate network connectivity. Authors with a large number of subscribers who each has only a handful of friends are in a better position than authors with a small number of subscribers who in turn may have a large number of friends.
- Influence rather than reach facilitates the diffusion of user-generated content, demonstrating the value of opinion leadership.
- Entertainment and educational values positively affect diffusion, whereas production quality did not matter.
- User ratings had an impact on a videos successful diffusion. Increasing the average rating by 1 star can lead to as much as 13.5% gain in diffusion rate.
- Younger users’ contributions are more likely to be popular.
- An author’s past experience and success positively affect diffusion of new videos.
We hope that marketers and public relations practitioners (as well as other content creators) will find this information useful in more effectively predicting the success of user-generated content and planning successful online campaigns. We also hope this study offers a stepping stone for more research on the topic.