If you ever feel like your Facebook ad results are too good to be true, they probably are. In the case of ThruPlay numbers, it’s likely due to the Audience Network Rewarded Video placement.
I’m glad I caught this quickly, but I guarantee that many other advertisers are wasting thousands and thousands of dollars, assuming they found some magic bullet that is giving them amazing results. Unfortunately, they likely won’t be too happy when and if they find out what is actually happening.
Let’s go through this from the beginning. In this post we’ll cover the following:
- What is ThruPlay?
- My campaign
- My results from ThruPlay
- What is Audience Network Rewarded Video?
- How to check placement distribution
- Why this was caused by Facebook Optimization
- What to Do?
What is ThruPlay?
ThruPlay is a metric used for measuring Facebook video engagement. A “ThruPlay” is reported whenever your video is played to completion — or for at least 15 seconds.
That final part (“or for at least 15 seconds”) is what makes this metric so misleading. “Thru” suggests someone watched the whole thing. Well, that’s the case if the video is only 15 seconds long or shorter. But if it’s 30 seconds or two minutes or 10 minutes — nope. It just measures the number of people who watched for at least 15 seconds.
Anyway, that the metric is misleading isn’t really the focus of this post. This metric is used for both reporting and optimization purposes.
I don’t want to get too lost in the weeds here, but I think it’s necessary to explain what I was trying to do. I used a 20-second video to promote a podcast episode. I used both a CTA button (“Listen Now”) and a link in the text to drive people to the episode on Spotify.
This is a new podcast and it’s a non-business project, so I wasn’t looking to spend much at all. Just trying to get it off the ground.
Since I was sending people to Spotify, measurement and optimization were a bit tricky. I don’t own the destination website, so I won’t know how many people play the episode or subscribe to my show. My primary measure for success in Ads Manager will be Unique Outbound Clicks since that’s the only way I can measure direct clicks on my ad away from Facebook.
I’ve never been a huge fan of Link Clicks optimization since quality can be low, so I created three different ad sets optimized in three different ways:
- Link Clicks
- Post Engagement
The ThruPlays optimization, of course, is what inspired this blog post.
My Results from ThruPlay
Keep in mind that my budget was extremely modest on this campaign, so I wasn’t expecting much. But I quickly noticed something weird with the ad set that was optimized for ThruPlay. The results were… practically impossible.
This is crazy. Nearly every view was a ThruPlay. The average watch time was 20 seconds (the entire thing). The total number of ThruPlays is higher (126%) of the total number of people reached. Frequency was just 1.31, so how in the world was this possible?
This math essentially means that nearly every person who was reached with my ad watched the entire video. I realize I make a damn good video (I really don’t), but this doesn’t seem realistically possible.
One more element: Of the 600 or so people who watched the entire video, only ONE resulted in an outbound click. So, my video was so engaging that people wanted to watch the entire thing. But, the entire point of the video was to get them to click and listen to my podcast episode. And yet, only ONE person did that?
This really didn’t make sense. It made me ask a lot of questions. I shared it with my Power Hitters Club – Elite community, and one very alert member suggested I look into the Audience Network Rewarded Video placement.
Audience Network Rewarded Video
Audience Network is how app developers monetize their apps with Facebook ads. They place ads inside their apps so that advertisers can target users with Facebook ads but away from the Meta family of apps.
Here’s the best way to explain Audience Network since it all made sense to me once I was exposed to my own ad the first time on Trivia Crack years ago. You’re playing a third party app that is probably free. It’s not a Facebook app. But your ads may appear there. That’s how those apps make money.
I actually remember seeing my ad on Trivia Crack. In fact, I clicked it accidentally — which is one of my biggest complaints about the placement.
But that was related to link clicks and landing page views. There is also a Rewarded Video placement. This is how Facebook defines it:
Rewarded video ads are a fullscreen experience where users opt-in to view a video ad in exchange for something of value, such as virtual currency, in-app items, exclusive content, and more.
Do you see how this placement is problematic? People are forced to watch these videos — not just a little bit, but the entire video — in exchange for something of value in the app. The user has no interest in the thing you are promoting in the video. They are only watching (loosely) to get something for it.
How to Check Distribution by Placement
Hopefully you’re aware of the Breakdown feature because it’s an immensely valuable tool. This is how you can breakdown you results by placement.
Click the “Breakdown” dropdown on the right in Ads Manager. Then select “By Delivery” and then “Placement.”
And there it is. The vast majority of my ad spend was used for the Audience Network Rewarded Video placement.
This would explain why the ThruPlay percentage was so high. These people were forced to watch the entire video. It also explains why there was only a single outbound click. These people are deeply engaged with an app (probably a game). They’re only watching the video to get something they can use for the app. Why would they click out of that experience?
They wouldn’t. And they didn’t.
How This Was Caused by Facebook Optimization
Of course, this is absolutely not what I wanted. I wanted people to watch my entire video — but only because the assumption was that if someone watched the entire video they’d be more likely to click to listen to my podcast. But that’s not going to happen in this case because it’s not normal.
This is one of the many problems with Facebook ads optimization. If you optimize for something, all Facebook cares about it getting you as many of those things as possible for the lowest cost. While that might be great for purchases, it’s often going to create issues for other actions.
I only want people to watch my entire video because I assume that these are normal people. And if they were normal people, watching an entire video would make them more likely to click to listen to my podcast. But that wasn’t happening. Why? Because this is the lowest quality possible of a ThruPlay.
The same thing happens if you optimize for Link Clicks or Landing Page Views or Post Engagement or just about anything else. Facebook does not care about whether the actions they get for you are quality actions. They only care that they get you as many as possible.
The problem, though, is that there are many ways to get these low-quality actions and Facebook’s systems will go straight to them if you tell them that’s what you want. This is one of many very specific examples of how it can be manipulated.
What to Do?
If you ever have this issue and you want to prevent your ads from being shown on the Audience Network Rewarded Video placement, you can.
Within the ad set, make sure to select Manual Placements.
If you want, you can just turn off Audience Network entirely under Platforms.
You can also choose to uncheck the Rewarded Video placement specifically if you want to keep the other Audience Network placements running.
So, I encourage you — if you get results that are too good to be true, don’t assume that you’re just an advertising god. Do a little digging. Think about why these results might be possible. Often, it’s a weakness in the way Facebook optimizes.
Have you seen these issues with Audience Network Rewarded Video and other placements?
Let me know in the comments below!
The post Audience Network Rewarded Video Placement and Inflated ThruPlay Numbers appeared first on Jon Loomer Digital.
Did you miss our previous article…