June 13, 2012 |
When it comes to marketing via video sharing on social media, choosing an influential person to “seed” a video is essential — but video quality isn’t all that important.
So concludes a new University of California, Davis, study of “buzz” marketing on YouTube. The research takes an important first step in looking at video viewership online, and in helping companies best use online campaigns to boost their return on investment (ROI).
“The identification of effective seeds (or primary authors) is not only key to the success of these campaigns, but also an important factor in estimating the return on investment from a manager’s perspective,” said Hema Yoganarasimhan, a professor at the UC Davis Graduate School of Management and author of the study.
“Seeding information in social media outlets through handpicked agents is now becoming a common strategy in buzz marketing campaigns,” the study notes.
Choosing the right seeding “agent” isn’t just about friends and followers, Yoganarasimhan found. “It’s not the number of people, it’s focusing on the right people,” she explained. “They need to ask who are their friends, and who are their friends’ friends — and how are they positioned in the network?”
She cites a Ford Motor Co. buzz marketing campaign for the subcompact Fiesta car in 2009 as a good example. Eschewing traditional marketing, Ford garnered 6,000 car reservations, 6.2 million YouTube views, 750,000 Flickr views and about 4 million Twitter impressions in less than a year. In what became a well-known campaign in the social marketing world, Ford selected 100 influential video bloggers, or “vloggers,” who agreed to try out a Fiesta for six months in return for blogging, tweeting and otherwise recording their experiences with the car on social media.
In her research, Yoganarasimhan looked at 1,939 videos that were uploaded to YouTube during November 2007. Among those, more than 1,800 were posted by authors who listed their friends.
“I visited the pages of these friends and obtained a list of their friends. So for each video, I reconstructed the social network of the author up to two hops,” Yoganarasimhan said. She then monitored those videos daily for more than a month for views, ratings, comments, sharing and other statistics.
While a close-knit community may be committed and loyal to a dispenser of information, that community may generate low video popularity in the long run, the study showed. That’s because people in a close-knit community don’t interact much with outsiders, resulting in few interactions with second- or third-degree “friends.”
The study found that while first-degree friends are important for initial marketing, second- and third-degree friends are essential for “viral” spread.
Finally, the study found that video ratings are important — but it doesn’t much matter if the rating is good or bad. Yoganarasimhan’s analysis showed that video quality, as measured by viewer comments and ratings, had little effect on viewership in the long run. However a video with any rating was likely to have more viewers than one with no rating.
YouTube is a good platform to study, Yoganarasimhan said, because it involves both social media and video sharing.
The paper: “Impact of social network structure on content propagation: A study using YouTube data,” is available at: http://www.springerlink.com/content/u243567n085537p6/fulltext.pdf
The research is published in the latest issue of Quantitative Marketing Economics.