If one video deserves to be viral, it surely must by Gangnam Style by Psy, the most watched video on YouTube with more than 2.2 Billion views, which recently hit the threshold of 2^31-1; the largest 32-bit integer commonly used in today's computer systems. 32-bit integers range from -2^31 to 2^31-1 since 32 bits (zeros and ones) can represent only 2^32 different numbers. Google used to represent the number of views with a 32bit integer but had to change thanks to Gangnam Style.
YouTube has a function for easy sharing of videos between friends. Let's examine the number of shares for Gangnam Style. The ratio of shares versus views is 0.11%. Are these shares sufficient for viral spread? First, this ratio is only slightly higher than for one of my own videos (0.07%), which I am pretty sure did not spread virally. To illustrate the non-viral nature, the following graph shows the daily views and shares for my video.
Instead of driving more views, the shares seem to be driven by the views. However, small changes in the ratio of shares can have a big impact on viral spread if we get close to a critical threshold. So we need to look closer at this critical number. What ratio of shares do we need for a viral spread?
Your friends and their friends form a network of people, a network with links between people that are friends, ultimately connecting all people on Earth. What is the condition of viral spread in such a network? The average number of facebook friends of U.S. users in 2014 was 350. Let's say 10,000 people noticed a specific video. If 0.11% of them spread the video, 11 will spread them to 350 friends each, reaching 11*350 = 3,850 people (actually, the number will be less since some friends overlap, but for my argument I just need an upper bound). In the next spreading stage, those 3,850 will spread it to 1,482 people. You see the number decays with every stage, which means that the video will not go viral. To go viral, the critical ratio of shares would be 0.29%, well above the value for Gangnam Style and any video I know of.
I like to add the caveat that the YouTube shares do not include all shares of the video and that the number of YouTube friends is likely lower than the number of facebook friends. These numbers might balance out, but I do not know for sure. However, I think that the percentage of our engagement with videos naturally regulates itself to prevent viral spread. Millions of creative people create compelling content every day. If our engagement would be so high to enable viral spread, we would be constantly flooded with shares, and this flood would reduce our enthusiasm to engage. So the level of engagement regulates itself.
Finally, you may ask if that is true, why did Gangnam Style get so many views? Apparently, the creators did not just rely on viral spread. The campaign was well planned. Four days before the main video, a teaser was released on YouTube, which instantly attracted a few hundred thousand views. That shows that Psy was no noname to begin with and the producers probably paid for advertisement. What followed was likely a clever engagement with the media to get the video into the international news. Of course, the video had to be compelling too, but on its own it would not have spread virally.
To conclude, creating compelling content is not enough, traditional marketing matters. You cannot just rely on viral marketing. I am curious about your thoughts and comments.
