Do YouTube views matter? For many they are a barometer of a film’s success on the channel, a simple way of counting how many people have engaged with a piece of work. Yet there is increasing debate about how accurate a marker a 'view' actually is. So Minneapolis-based creative shop Solve decided to put the metrics to the test, by seeing if they could make a four minute-long blank video go viral…
YouTube views appear an easy signifier to use to judge the popularity of a video – I’ve cited them many a time in an article to illustrate the success of a piece of work. But for marketers they have accrued an added power, becoming key data to determine whether an ad has resonated with its audience. Solve’s project set out to challenge whether views are truly a good measure of engagement, or whether they can be fudged.
“As an agency that values both creativity and results, we wanted to provoke marketers and industry colleagues into looking past views,” says Eric Husband, creative director – integrated at Solve. “Another prompt was pure curiosity. Should we be so obsessed about views? I suppose we could have written a nice little opinion piece. But we wanted to go deeper, more factual. Which honestly made it more fun.”
The premise was simple: the team loaded up a blank video, with no sound or motion. They then used YouTube’s TrueView In-Stream advertising to promote it, spending $1,400. From that investment, the video gained 100,000 views.
Watch the video in all its glory here
“Per YouTube’s advertising policy, we were only charged if a viewer watched at least 30 seconds of our four-minute blank white video,” explains Neil James, digital strategist at Solve. “All viewers had the option to skip the video after five seconds. As a result of our promotion, we were able to generate over 100,000 views for only a $1,400 investment – or 1.4 cents per view. The ad itself was served 227,819 times, meaning that the view threshold was reached nearly 46% of the time.”
Other surprising stats followed. On average viewers made it through 61% of the video before skipping it, and 22% made it to the end. There are a number of possible theories as to why this occurred. Perhaps viewers saw the film as a kind of video version of John Cage’s 4’33’? Or they were curious at the lack of activity, wondering if something interesting might eventually happen? (The video was originally posted with no title or description, which might back this idea up.)
Or, and this is most likely, these views happened by accident. “Although it’s impossible to know, it’s unlikely our blank video was actually viewed 100,000 times in the traditional sense,” says James. “In reality, most of the views were probably attributable to inadvertent plays – pre-roll streaming in the background while multiple browser tabs were open or people mistakenly believing their intended video was loading. In either case, the deliberate intent needed to legitimise the view as a proxy for video quality is conspicuously absent.”
Views have been a source of much debate of late. After Facebook trumpeted that it was now streaming more video than YouTube, “professional YouTube creator” Hank Green wrote a powerful blog post, laying down his issues with the way the site is set up, which opened with the statement that “to be able to make that claim, all they had to do was cheat, lie, and steal.”
While Green is concerned about Facebook’s lack of interest in creators in the way they share films on the network, Solve is more focused on the over-emphasis that marketers give to views and in turn how this affects the work that gets made.
The Blank Video Project shines a light on some of these issues. “It showed that YouTube views alone cannot measure the power of content,” says Husband. “We think there are better stats such as likes and shares on a per view basis – these are also difficult to inflate or game. We found that a video can rack up a ton of views with a low level of likes [The Blank Video Project notably has no likes or shares]. Which means the video didn’t necessarily go viral, the views were just purchased. We can put our weight into other meaningful stats too – from clickable calls-to-action to conversion.”
“It’s all too easy for brands to forget that content consumption is a means to an end – not the end itself,” continues James. “Branded content that fails to favourably change the attitudes, beliefs and behaviours of target audiences cannot be considered a success regardless of how many millions of views were accrued.”
YouTube has declined to comment for this story, but did point me towards the fact that YouTube has been using watch time, not views, as its key metric since 2012 (more on this here).
And Husband stresses that Solve has no particularly bugbear with YouTube itself. “YouTube is a great channel. We don’t think it’s YouTube’s fault that some marketers are fixated on views. We’ll continue to use the channel, but we certainly won’t look at views as the only measure of success. That said, our blank video has now surpassed 154,000 views. But who’s counting?”