How Many Views Does It Take To Go Viral?
What a wildly complicated and time sensitive question. I’ll give you a concrete, numerical answer at the end of this diatribe, but first, I feel it’s important to establish some background. In the 30 year history of internet video, nothing has changed more dramatically than the goalpost of “virality.”
Since November of 2020, the most watched video on YouTube has been Baby Shark Dance. At the time of writing, Baby Shark has amassed more than 16 billion views. For context, that’s about twice the population of the earth. TWICE THE POPULATION OF THE EARTH.
So, to be on top of the leader board, you need everyone on earth to watch your video at least twice. No pressure, right?
The history of the “most viewed video” crown tells a more complicated story. When Baby Shark Dance topped the leaderboard, it did so with just over 7 billion views. For those playing along at home, that’s less than the population of Earth.
So, in less than 5 years, an already astronomical goalpost has been pushed back 230%. Ready for an even crazier number? PSY’s Gangnam Style was the first YouTube video to pass 1 billion views. It did so in December of 2012. That means in less than 13 years, we’ve kicked that goalpost back 1600%.
Just to cement the statistics here, I’d like to give you one more point of reference. If you were on the internet back in 2009, the title “Charlie Bit My Finger” should be familiar to you. The home video of a teething toddler became the most viewed video on YouTube in October of 2009. Charlie and his brother claimed the title with around 129 million views. It’s been 16 years since 2009 and we’ve moved the goalpost back 12,400%.
So what the hell happened? The population of the Earth hasn’t changed that dramatically. We haven’t found a way to slow the passage of time. Across these 16 years, the number of internet users has grown by 318%, but that still doesn’t even nip at the heels of 12400%. If there’s not a glut of new users to generate these views, where are they coming from?
Change in Digital Infrastructure
There’s no way to know exactly when the first video was uploaded to the internet, but it was sometime in the early 90s. Back then, there were only about 25 million internet users worldwide. It’s important to note that the internet of the early 90s wasn’t as interconnected as the internet of today. So, while there were ~25M folks online, you would likely only interact with a small portion of them. Perhaps you were a regular on a forum or chat board that expanded your reach, but the concept of “upload content for everyone to see” just didn’t exist.
All the way back in 1892, a Russian botanist named Dimitri Ivanovsky discovered that tobacco sap, filtered of its bacteria, could still infect healthy tobacco plants. Then, a Dutch botanist named Martinus Beijerinck called the infectious substance in the sap a “virus.” These early experiments established the defining trait of viruses: they spread.
I’m not going any deeper on the history of virology because we’ve identified their key feature. Viruses spread, so for something to be “viral,” it has to spread. That means “virality” shouldn’t be a measure of “how much,” but instead a measure of “how far.” But, as noted above, the internet has gotten smaller. Where there used to be hundreds of niche websites with their own communities, there are now a handful of hub sites. So if you can’t measure “how far” by distance traveled, how do you measure it?
What Counts as a “View”
In 2005, one of YouTube’s co-founders, Jawed Karim, uploaded the platform’s first ever video called “Me at the zoo.” For the next four days, “Me at the zoo” sat alone on YouTube and not a single person watched it. With his first video still at zero views, Karim uploaded his second video, one of a jet plane taking off. On April 27th, 2005, this second video received its first view, and the rat race began.
Sure, other videos had been uploaded to the internet before, but they weren't displayed with a view count. As YouTube solidified itself as a hub site in the new digital infrastructure, the view count leaderboard became highly coveted. Seven months after YouTube went live, Anthony Padilla & Ian Hecox uploaded a music video of the Pokemon Theme Song to their channel called Smosh. Five months after that, Smosh’s video became the first upload to reach 100 views.
As view counts reached triple digits, creators began to ask a very important question: “What counts as a view?” This question rose from the desire to quantify audience size. If a video earned 100 views, does that mean 100 people watched it? Or could one person watch it 100 times? Unfortunately, that wasn’t a question YouTube was capable of answering. In fact, they couldn’t even guarantee that one person had watched a video one time. What if a user clicked play, got up and walked away? To YouTube, that was a view, but anyone with a brain can tell you it shouldn’t be.
After 20+ years of practical experience and technological advancement, YouTube has tightened their definition of a “view.” The general understanding is that a video must be played for 30 total seconds before YouTube counts a view. But, both on YouTube and YouTube Shorts, there are millions of videos shorter than 30 seconds. For those videos, YouTube counts the view as soon as the video begins. Every video platform has a different definition of a view; you can read about some of them here. Fair warning though, platforms are not under any obligation to be honest about their definitions, and most of them are not.
What none of these definitions account for, YouTube’s included, is the viewer who got up and walked away. YouTube uses skippable ads in a similar way to Netflix’s “Are You Still Watching” screen, but that can only demonstrate a single instance of engagement. I can’t be the only person who has put on a background YouTube video, gotten up to do something else, heard an intrusive ad, walked back to my TV to skip the ad, then immediately gone back to what I was doing. My point is that YouTube (and any platform, for that matter) can only measure when a video is played, not when it’s actually viewed. This isn’t just semantics, it’s the difference between someone walking by your flier and someone picking up your flier to read it.
The Need To Sell Ads
If I haven’t made it clear, it’s incredibly difficult to accurately quantify a view. But here’s the thing, if you can’t accurately quantify something, you can’t sell it. The infamous “pivot-to-video” of 2015 created a ravenous hoard of creators, itching to buy views. To satiate the masses (and to line their pockets), digital hucksters hustled to clearly define a “view.” They found a solution in a concept from 20 years prior, the “impression.”
In 1994, communications megalith AT&T purchased the internet’s first banner ad on hotwired.com. One year later, the CPM (Cost Per Mille) pricing model became standard, and advertisers began to charge in increments of 1000 impressions. The Internet Advertising Board (IAB) began to standardize audience measurements online in 1998, but an impression has never meant more than a single instance of visibility.
In 2015, with the CPM model fully tenured in internet advertising and creators hungry for growth, executives began loosening the definition of a “view” to line up more with an “impression.” Suddenly, view counts began to skyrocket across multiple platforms. More than anywhere, videos uploaded to Facebook seemed to generate views at an alarming rate. Legally, Facebook had nothing to do with this. In fact, they paid out a $40 million settlement in 2019, a thing they totally wouldn’t have done if they actually inflated view numbers. Definitely-not-guilty corporations pulling the strings aside, it’s easier to sell views when you can guarantee more of them. The easiest way to guarantee more views is to lower the standard of what’s considered a view.
The Rise of Fake Views
Are you feeling disillusioned yet? So far, we’ve established that the goal post for virality has become unfathomable, the infrastructure of the internet doesn’t allow us to measure what virality truly means, the definition of what constitutes a view is an ever moving target, and that platforms are incentivised to mislead you. Still with me? Well, strap in, I’ve got one more incredibly bleak thing to talk about: click farms.
If you’ve been on social media since, oh let’s say 2016, you’ve likely heard tell of “fake views.” For those of you who aren’t chronically online, I’d bet that the phrase “fake views” is a little confusing. How can someone fake watching something?
Well, in most cases, it’s not actually the views that are fake, but the people watching them. Imagine a person who is paid an hourly wage to watch and interact with videos on TikTok or Instagram. In an 8-hour work day, that one person could easily scroll through 400+ videos, but would they actually watch any of them? Now imagine we give that one person 50 phones and let them scroll on all of those phones simultaneously. Now our one person can scroll through 20,000 videos in an 8-hour day. To TikTok or Instagram, that’s a solid 20,000 views. But to a small content creator trying to build an online community, that’s just one person.
This operation has existed as a business model since at least 2007. Here’s an article from CNN (with pictures!) investigating a click farm in Vietnam. Teams of underpaid workers in small countries all over the world are manning walls of smartphones, generating millions of social views. Are these views real or fake? Well, to the platforms selling views and engagement through post boosting, these views are as real as they get.
Unfortunately, the internet is structured around large numbers. In the age of algorithmic discovery, content with high view counts will continue to get views. If you’re paid per view on social, that’s terrific, but most users will never reach that category. Most users will continue to yearn for a higher view count, but social media is a rigged game and The House always wins. The House isn’t the content, The House isn’t the creators, The House is a hoard of executives who have created a way to sell something that doesn’t actually exist.
So How Many Views does it actually take to go viral?
I asked my newsletter subscribers and my TikTok community. Here are some of the answers I got:
300 +
1,000+
10,000+
100,000+
1,000,000+
1,500,000+
I also got two non-numerical answers, and I think they’re both on to something:
“A video is viral if it gets 15x the average amount of views that creator receives”
“A video is viral if I see it on a second platform.”
I think both of these responses have merit. They tie back to the original “how far” definition of virality. If a creator averages 100 views per video, and suddenly a video gets 1500 views, you can safely assume that video has traveled some distance. If you upload your content to one platform and someone reshares it on another, that’s empirical spread. Neither of those answers are absolute, but I think they answer the question.
My personal threshold is lower. I consider content to be viral when it has more impressions than its poster has followers. If you have 30 followers and 100 people watch your video, congratulations, you’ve gone viral. That may seem low, but I think it would surprise you how infrequently that actually happens. In fact, the average subscriber-to-view ratio on YouTube is only about 7%.
For those of you who want a more concrete answer, my numerical threshold is the letter K. When a piece of content has so many views that they’re quantified with a K, I consider it to be viral. On most platforms, view counts switch from exact to “#K” at 10,000. But even this is subjective to the creator, and that’s just the reality; “virality” is subjective.
If you enjoyed this blog, consider subscribing to my newsletter! One email a week and no spam, I promise. It will keep you updated on when I write more things like this and all of the other content I produce. Subscribe for FREE. 