- A video view doesn’t mean someone paid attention to your video.
- Be careful who you retarget.
- It's not what you want; it's what your audience wants.
Let’s say you just bought a Ford Shelby GT350 Mustang. The body of the car looks mean, ready to blow zero to 60 in 4.3 seconds.
But the body doesn’t matter all that much — it’s the engine, hidden underneath that body, that makes all the difference.
Replace that 536 horsepower V8 with a Ford Taurus engine, and you’re not going anywhere in less than five seconds.
The same applies to Facebook video views: Don’t fall in love with the massive amount of Facebook video views you’ve been receiving unless you understand what’s going on under the hood.
[Tweet “Don’t fall in love with the massive amount of Facebook video views you’ve been receiving.”]
I’m sure you hear the gurus out there, claiming that they can get video views for 1 to 2 cents a pop. That sounds great, and might be true, but what is a video view, and more importantly, how can you get your money back from your video campaigns?
What is a video view?
Facebook requires a video to run for at least three seconds for it to count as the user viewing the content.
Even if it counts as a view, did the viewer take away anything from your video?
Think about your own browsing habits: Do you ever notice videos running on the screen, and you couldn’t care less? Do you ever take the time to pause running video?
Videos viewed on mobile start running once half of the post is on the mobile screen, so it’s totally possible to consume the content above or below them while the video plays. Videos on a desktop don’t run until they are 100 percent in your feed, but the bigger screen still allows you to ignore them.
If you happen to be retargeting using the custom audience for video views — don’t. The vast majority of those people don’t care to see another ad that takes them deeper into the sales cycle.
The vanity metric that Facebook puts on its views sucks — but Facebook videos and the behind-the-scenes metrics are amazing for dialing in on people who are interested in your business.
So how can you legitimately measure how interested people are in the videos you put out, aside from the standard metrics (shares, likes, comments, etc.)?
How to measure legit views
First, use the standard video view metric provided by Facebook. How much did the user watch — 25 percent, 50 percent, 75 percent, 95 percent or the whole shebang?
This is where you start to see who is truly interested in your content. It’s also where the custom audience comes back into play: You want to retarget someone who watches 95 percent or more. This person is engaged and likely to be interested in what you have to say (or sell) down the road.
It’s important to note that the addition of captions has been huge in keeping users interested, as the volume isn’t present until the video gets clicked.
[Tweet “You want to retarget someone who watches 95 percent or more of your Facebook video.”]
This brings me to the hacked metric I use to measure our videos’ success: cost-per-click to play.
Our cost-per-click to play tactic
Sort by video engagement in your ads manager, and then break it down by action type and select video view type. These are the users who cared enough to hear what you have to say.
I take my entire budget for the video and divide that by how many people clicked to play (and hear) what was being said. These clicks are recorded under the “Performance and Clicks” filter, but if you want to know the cost-per-click to play, you need to do it manually.
One of my shorter, two-minute videos had many views all the way to 95 and 100 percent, which looks great on the surface — but only three people actually clicked to play discounting my initial value of the video.
This is the difference in wasting valuable time and money on ads and content that isn’t as relevant as you might think. It’s not what you want — it’s what the audience wants.
Adam Stephens is a branch partner at The Stephens Brotsky Group at Primary Residential Mortgage Inc. Connect with him on LinkedIn or Facebook.
