A whopping 300 hours of video are uploaded to YouTube each minute, which is a threefold increase since May 2013, reports ReelSEO. And while YouTube is currently the most dominant video hosting service, other platforms are experiencing similar trends as well. In fact, the amount of video content being uploading to Facebook has been trending steadily upward since the beginning of 2014 and could surpass YouTube in this year, according to research from SocialBakers.
This trend presents a number of issues for both viewers and media providers. The former require a fast and reliable means of discovering content that correspond to their interests. In turn, the latter require an efficient and dependable means of providing content that their users will like–whether these means are provided by hosting services like YouTube or not.
Why Searching and Browsing Are No Longer Sufficient
There is an overwhelming amount of video content out there for viewers to wade through. Of course, it’s impossible for any one person to watch every video themselves, but it’s becoming increasingly cumbersome for people to even find content they want–unless it appears at the top of a search, but this creates other issues.
Virtually any search will turn up a plethora of results too numerous to go through one by one, so people tend to select the first few results, which limits engagement. People will find the video or type of video they’re looking for, watch it and leave–unless they have easy access to other videos they are likely to prefer. This is why YouTube has invested so much time and energy into its recommendation engine. An effective recommendation engine can reduce the effort a user needs to apply in finding content she likes. Of course, YouTube and other video hosting services, broadcasters and publishers want users to find content of their choice as this increases engagement (which often translates into increased revenue). But how does the YouTube recommender system work and what are its utilities and drawbacks for the companies and people who use YouTube to host their videos?
The YouTube Recommender System
The YouTube recommender system’s goal is to provide high quality, relevant and fresh recommendations to its users. These recommendations are also intended to be diverse in order to reflect the wide array of content on the site–the system is designed not only to provide video suggestions that are relevant to whatever the user is watching at a given moment, but relevant to the user’s overall tastes. Recommendations have been shown to account for around 60% of clicks on the YouTube homepage and produce a click through rate (CTR) 2.7 times that of the Most Viewed videos. To learn more about the YouTube recommender system click here.
What About Media Providers Who Can’t Rely On YouTube?
YouTube is an excellent hosting platform and its recommender system can potentially generate exposure for media providers. However, if these media providers have their own sites that they want to keep viewers engaging with, YouTube won’t help.
Media providers don’t need to have the resources and expertise of Google to implement a highly effective recommender system on their own sites. Indeed, companies such as Trouvus, specialize in developing top-shelf recommendation engines for media providers looking to significantly improve engagement, retention and ultimately revenue. As long as media providers have a sufficient amount of content for the recommender system to parse, providers can use out-of-the-box solutions that are comparable to the YouTube system in term of effectiveness. In fact, our own internal tests on our system have yielded better results than those documented by YouTube.