Social video platforms are in desperate need of standards when it comes to video view definitions. When I talk about standards, I don’t mean it in the sense of “Who hired Frank? He really laid an egg with that campaign report.” I mean it like this:
Specifically, we need a standard for how social video platforms measure consumption. Such a standard would make genuine comparisons of popularity across platforms possible at the video level, creator level, or network level.
The problem is, no such standard exists that is publicly displayed by any social video platform today. “But views!” you say. And how!
Unfortunately the state of what constitutes a view varies so widely that it makes any apples-to-apples comparison of consumption across platforms a bit foolish. See this helpful graphic from Paladin below for view definitions as they exist today across the major social video platforms (Facebook, YouTube, Snapchat, Instagram, Vine, Twitter, Periscope).
You might have noticed there is not much consistency. But one common thing that can be gleaned from this data is the intention behind defining a view. Vaguely, it is a video start that has accrued enough watch time to be considered significant.
This leads into the biggest point of debate: what is significant when we’re talking video consumption? If you’re deep in the data weeds like I am, you’ve probably had some kind of insane inner dialogue like this:
What would be a significant and transparent view… Let’s see… 30 seconds? That’s what YouTube uses to define an advertised view… That sounds good. But what about hyper-short-form content on Snapchat or Vine and the content is only 3.6 seconds long? Ok let’s use a percentage then… But what percentage? 10%? That’s far too short for a 3.6 second video (0.36 seconds)… And far too long for a 1 hour video (6 minutes)… Maybe a sliding scale… A higher watch time percentage required for short content and a lower percentage for long content… But then the view count isn’t transparent… I should really make some friends soon.. I wonder what’s for dinner..
Here are a few other fundamental weaknesses of the view count:
- Views treat a viewer equally if they watch a piece of content from beginning to end, or if they only watch the minimum amount required to add to the view count.
- Views are treated equally across content with varying lengths (in most cases).
- Views do not account for the actual number of unique individuals who consume a given piece of video content.
My question is, why are we basing an entire industry around a metric with more holes than swiss cheese? The only answer I can come up with is that we shouldn’t.
In my opinion, you can boil down consumption to 3 genuinely meaningful metrics to paint a full picture of a video’s popularity.
#1: Minutes watched
How many total minutes of watch time has this piece of content accrued from all viewers over its entire lifetime.
Time truly is the ultimate measure of how popular anything is. Video influencers, TV and film producers, app developers, bloggers, etc are all competing for one thing: your time. If we were to reduce this list of 3 metrics to 1, this would be it. Just show time instead of views.
#2: Unique viewers
How many individual people actually watched the video.
Did 100 people watch a video once, or did 1 person watch it 100 times? This is an important piece of context when trying to measure the real-world popularity of a thing. Uniques are a challenge to accurately measure given that an individual can watch from multiple devices, and multiple people can view from the same screen. But I’m confident the product wizards working for these platforms can put something together that’s good enough for government work.
The average percentage of viewers who consumed the content for any given second of it.
This is increasingly crucial as live video becomes more prevalent. Where did viewers join the broadcast? Where did they abandon a broadcast and what made them do it? If they are watching a non-live video, what part of the content did people skip or watch over and over?
While this data is available privately on YouTube and Facebook today, I feel it would push the industry to make stronger content if it were available on more platforms, publicly visible, and available via API without restriction.
Here is an example of how I think Minutes Watched, Unique Viewers, and Retention data could all be presented neatly on a typical widescreen video player.
Data appears on top of the video frame when the player controls are in use. Summary data is nested at the top while second-to-second retention is overlaid as a graph. The video frame is darkened to help the data pop but is still visible.
Keep in mind that the three metrics we have included are useful for all types of media with a beginning and an end, not just video. They are important data points for podcasts, music recordings, audio books, live broadcasts, etcetera.
Obviously there are significant differences between the social video platforms mentioned here. But in this time of rapid growth in digital video, we would do ourselves a favor by separating the apples from the oranges.
If you’re looking for tools to help manage and scale your MCN, click to learn more about our Network Management Suite.
If you’re looking for tools to help find influencers for brand campaigns and network recruitment, click to try our Talent Locator for 1 month free.
If you’d like to track and monetize your IP across YouTube, click to learn more about our Rights Management Suite.
Kudos to Jennifer Dagdigian for neatly crafting the view definitions infographic above, and to Silviu Runceanu for turning my wacky video player sketches into something that actually makes sense.