Stay on Top of Enterprise Technology Trends
Get updates impacting your industry from our GigaOm Research Community
Want to be the first kid on your block to forward the next viral video to all his friends? Well, a startup called PopScreen thinks it can help, with the launch of its “prediction engine” for determining which videos are going to hit it big.
PopScreen’s prediction engine works by tracking more than 10,000 sources across the web, such as YouTube (s GOOG), Vimeo and Dailymotion, as well as networks like Revision3, Babelgum and FunnyOrDie, to chart the trajectory of a video. By looking at how many views a video gets in a certain period of time, as well as tracking how linking from key influencers affects interest in a certain video, PopScreen’s algorithm can determine which videos are most likely to get popular.
You can see the algorithm at work in the following example, where PopScreen was able to predict the virality of the BP Spills Coffee video:
PopScreen, which was founded in 2008, originally started as a video bookmarking service, but the founders decided to use their search engine backgrounds to create the company’s prediction engine instead. The startup is privately held and has a staff of about 10 employees, with a home base in Houston, Texas.
While PopScreen’s website currently only shows videos that it expects to get popular any moment now, it hopes to roll out a separate page for videos that are “on its radar,” with predictions for the likelihood of certain pieces of content hitting it big. PopScreen co-founder and COO Glenn Gutierrez said in a phone interview that the company could leverage this data to provide a subscription service to allow networks and content providers to monitor which videos are about to get hot.
Related content on GigaOM Pro: New Use For Web Stats: Finding Hot Markets, Offline (subscription required)