Creating a viral video is the hope of nearly all online video creators. For some, virality is a primary goal; for others it’s a welcome byproduct of creating engaging content and effectively marketing it. Any serious content creator understands that things like optimized metadata, attractive thumbnails, pop culture tie-ins, controversial subject matter, and (perhaps above all) support from influencers, can make a piece of content a good candidate for virality. Some companies claim to sell virality, but their methods are unknown, their results hidden, and their price tags hefty.
A Viral Video Case Study: 5 Million Views In One Week
Considerably less well known are the actual mechanisms behind the process of organic virality. People vaguely understand that attention from sites like Reddit and Buzzfeed are positive, but specifics are rarely disclosed. How many views does a site like Reddit typically refer? Where, exactly, do all the millions of views that a viral video gets come from? And how does a creator (or marketer) help the process along? To help chip away at those questions, what follows is a breakdown, using real-time recorded notes and retrospective YouTube analytics, of a viral video entitled Elephant Attacks Safari Jeep. The JukinVideo YouTube channel’s most viewed video of 2013, Elephant was posted in late August and promptly went viral, gaining nearly 5 million views in its first week.
Three Steps To Going Viral: The Initial Push
There’s a myth among smaller YouTube channels that big subscriber numbers are instrumental in virality. Our channel had 188,000 subscribers the day this video was released – not huge, but nothing to scoff at; those subscribers accounted for merely 9,500 views on the first day. The subscribers are helpful to get a small handful of views right off the bat, but they do little in the way of virality.
One effective tactic in this early stage, regardless of subscribers, is to get other YouTube channels to ‘Like’ the video. Approach YouTubers that post content similar to your own, and offer to exchange ‘Likes’ to ensure your video gets into the feeds of those channels’ subscribers. The (relatively) new ‘Community’ tab in the YouTube analytics can help identify outside channels/users that regularly engage with your content. "Elephant" benefited early on from ‘Likes’ from several channels with fairly substantial followings.
Three Steps To Going Viral: The Early Stages
Many people hear the word influencer and think of somebody like Katy Perry who has millions of Twitter followers. Sure, if Perry were to tweet your video it would get a lot of views. However, basing a strategy around targeting Perry isn’t really a realistic one, let's be honest. Another kind of influencer, the much more reasonable kind to target, is the small blogger that is seen as a leader in a particular field. Early influencers for "Elephant" were smaller blogs like SayOMG and a half-dozen others that typically post funny caught-on-camera videos with low view counts.
Every niche has blogs that cater to it; we make it a priority to identify and alert these early influencers when we have something of value to offer. Key for us in those relationships is that we do not spam them with every piece of content we put out; we wait for the videos that have big potential.
To be clear, these sites DID NOT refer a high volume of traffic. All told, they provided less than 65,000 combined views on "Elephant", which represent less than two percent of the total view count. Their value is not in the quantity of viewers they send, but in the quality; moderators of the bigger, more popular sites look to them for content. Indeed, Gawker’s viral content extraordinaire Neetzan Zimmerman recently told the Wall Street Journal that every morning he scans a feed of over 1,000 smaller sites for content to cover on Gawker. Getting the attention of a monster site like Gawker is the fastest way to virality.
Three Steps To Going Viral: The Next Steps
In the case of "Elephant", Yahoo! was the prime mover. As momentum had been building on smaller sites, GrindTV Outdoor, an affiliate of Yahoo!, published a nice article surrounding "Elephant", and embedded the video. Seeing how well the article was received on GrindTV, Yahoo! added it to the site’s front page slider. This was monumental; a video with roughly 150k views from small blogs and YouTube traffic gained another 300k from Yahoo! alone in just a few hours.
As an aside, only a handful of sites in the US can refer such a high volume of traffic in such a short window of time: Yahoo, Gawker, and Reddit certainly, and probably HuffPost, BuzzFeed, and 9gag to a lesser extent. I’ve left out a few others, to be sure, but not many. Following the Yahoo! story, several other mid-size sites like AOL’s Paw Nation picked up the story, referring an additional 100k or so views, collectively.
To this point (~28 hours after publishing), the video had been posted on more than 25 external blogs and websites, many of which we pitched the video to, others of which picked it up on their own.
The final event that cemented "Elephant" as a viral smash came from none other than YouTube itself. By midday on day two, YouTube noticed it had a hit on its hands and selected "Elephant" for its homepage. This was by far the biggest break for the video as the YouTube homepage referred a whopping 1.3 million views over a 48-hour period. If the Yahoo! homepage instigated virality, the YouTube homepage sealed it.
Broadcast Media: The Wildcard
The biggest unknown in this video’s success was television. Within the first 72 hours of the video’s release, Jukin Media had licensed the video to three nationally televised shows. While we’re certain that this draws positive attention to the original YouTube video, there is no specific metric within YouTube’s analytics or elsewhere that can identify exactly how many views a TV spot will send. We rely on close observation and analysis to piece the puzzle together.
Since our core business is media licensing, and our clients’ YouTube videos are shown on TV on a daily basis, we have at our disposal a mountain of self-collected data on the topic. The data indicate that the effect of TV on a viral video is highly variable; that is, it’s sometimes significant but other times negligible. We can say with certainty, however, that a large website like Yahoo! refers many times more traffic than even a primetime, network TV spot.
Interesting Facts From "Elephant" analytics
- More than 350,000 views came from foreign blogs embedding the video; advertiser value aside, a view in Kazakhstan is just as valuable for virality as a view in the US. Takeaway? Don’t forget foreign sites when promoting your video.
- Bucking traditional thinking on virality, less than 3% of "Elephant’s" views came from social media. Social sharing is one avenue to virality, but clearly not the only one.
- It received 5.1 million views in its first two weeks, and has gained nearly 400k views in its subsequent ‘long tail’ phase
- Reddit accounted for only 500 views; studying Reddit traffic to our videos over the years, the data suggests that the social aggregation site refers roughly 10 views per ‘upvote’ on a YouTube video post.
The case of "Elephant" is not meant in any way to be inclusive of all viral videos. There are many ways that a video can go viral. Ideally, the small nuggets of information that can be gleaned from this video contribute to the larger understanding of virality and can be useful for creators’ and marketers’ future video strategies.