GREG:
Yes. And we’re walking around it, but I think we are going to have to go right to measurement. Because if you aren’t measuring what matters, then you never know if your old ad has worn out its welcome or this particular message just didn’t resonate on social media.
Avoiding Vanity Metrics
So, there are a couple of vanity metrics that the platforms give you—and they’ve been giving you for years—that you need to question or you get into trouble. Are we ready to get into trouble?
MATT:
Let’s do it. I like starting with the vanity metrics. Let’s start with what not to do. And then we’ll get into what to do.
GREG:
Well, one of the metrics that you’ve got to question is “Followers.” Now, YouTube calls them Subscribers. Facebook initially called them Fans and now calls it Page Likes. But Instagram, Twitter, and Linked In call them Followers.
So, different platforms have different names for it, but the number doesn’t measure how many people you’re reaching. Why? Well, all of the social media platforms have adopted algorithms because they can’t show everybody everything posted by an account that they followed, liked, or subscribed to once upon a time. They show you some of them.
So, those old numbers—your Followers, Page Likes, and Subscriber—were devalued. As we mentioned earlier, the average organic reach on Facebook is 5.2 percent, if you’re lucky. And it’s around 1.0 percent on Instagram and only 0.045 percent on Twitter.
It’s an even harder thing to get real metrics around how many people you reached when you run social media advertising campaigns across multiple platforms. What happens if the same person sees your video ad seven times? Do your metrics tell you that you reached seven people, or do you realize that all you reached was one person who got very annoyed?
MATT:
Maybe they were a big fan.
GREG:
Yeah, right. So, one of the vanity metrics you’ve got to question is Followers.
Another one that I see people getting into trouble with again and again, is “Views.” People keep saying, “Look how many views my social video got.”
Well, how do you define a view? Because YouTube defines it as up to 30 seconds long. Facebook and Instagram define a view as anything over 3 seconds long. And TikTok says, “As soon as the video starts playing, that’s a view.”
So, a “view” is not a “view” is not a “view.” You’re comparing apples to kumquats and grapes. So, that’s another vanity metric that you need to question.
MATT:
There’s one more vanity metric. And that is “Impressions” on a lot of platforms.
When I am teaching analytics to marketing teams, I will give them a multiple choice option, “Define an impression,” and I give them five options.
I get a pretty consistent distribution of responses. That tells me the team doesn’t have a single definition of an impression. After I point this out, I ask, “What made you select that?”
I point out that as a human, we are thinking of an impression in a human term, which is I physically saw it.
However, several years ago, the Media Rating Council (MRC) created a standard definition of viewable ad impression: “At least 50 percent of an ad must be in view for a minimum of 1 second for display ads or 2 seconds for video ads.”
Here’s how YouTube and Facebook calculate an impression for a video ad: When “50 percent of the ad’s pixels” are viewable for “2 continuous seconds.” Facebook’s definition of an ad impression is anything greater than 0 pixels that is shown for more than 0 seconds!
So not everyone in the industry adopted the MRC’s viewable ad impression measurement guidelines. That’s a big problem. Because when consistent advertising viewability standards aren’t used, marketers may come to inaccurate conclusions about the effectiveness of their ads, which could lead to misguided buying decisions and suboptimal results.
And when I’m reporting up the ladder, when I say, “a viewable ad impression,” what are they thinking? They’re thinking less and less in digital terms and more and more in human terms.
Viewability has proven to be a valuable metric for brand campaigns. Research has found a consistent relationship between how long an ad is viewable and increases in awareness and consideration.
Other research has found viewability is as important for performance outcomes, such as conversions, as it is for brand outcomes. So, it may sound obvious, but video ads that have a chance to be seen are the ones that drive business results.
So, it’s important to understand the vocabulary and have an organizational definition so that we all know what it is we are buying, reporting, and making decisions about.
GREG:
Yeah. And those are things that you do not want to take for granted and just use because they’re convenient and available. You really need to think about and adopt metrics that matter. So, maybe we should begin to segue to, “How do I get out of this predicament?”
MATT:
So, what do you measure?
GREG:
Well, let me tell you about an organization called the Global Video Measurement Alliance (GVMA). They were formed in 2019.
They are developing cross‐platform standard metrics for social video, including ones that are significantly better than “Followers” and “Views.”
For example, the GVMA developed a new metric called de‐duplicated “unique viewers.” Instead of using Followers, Page Likes, or Subscribers, you can use this new metric to measure reach more accurately. It also enables you to leverage the power of de‐duplicated viewership to make better content and media decisions on both platforms.
Another new metric developed by the GVMA that you’ll want to adopt is “quality views,” which uses a 30‐second view as a qualifier. Any video shorter than 30 seconds is counted only if it is watched to completion.
Now, these are relatively new definitions. They have not been widely adopted as industry standards yet. But I would certainly encourage social media marketers who are serious about solving the problem of vanity metrics to pay attention to the GVMA. They are tackling the biggest problems of social video audience measurement.
And it’s worth noting that Tubular Labs, which helped to establish the GVMA in 2019, expanded their Audience Ratings suite in 2021 by adding ecommerce measurement capabilities. These provide a unique view into how product‐related social video viewing influences online sales through Amazon.
Measuring Brand Lift and Engaged‐View Conversions
MATT:
We both think that “Impressions” is a vanity metric. So, what should you use as a key performance indicator (KPI)?
GREG:
There are different answers to your question, depending on whether you’re launching a brand‐building campaign or running and performance marketing campaign.
For brand‐building campaigns, YouTube, Facebook, Instagram, Twitter, TikTok, Pinterest, and LinkedIn all offer “Brand Lift” studies to advertisers. These measure the impact of video ads on metrics that matter like brand awareness, consideration, and purchase intent.
They work by dividing the people into randomized test and holdout groups and showing your video ad to the larger test group, but the smaller holdout group doesn’t see it. Then, they conduct surveys comparing the performance of these groups over time.
To measure the impact of organic social video campaigns, you can conduct one brand lift study before a campaign is launched and a second one after it ends. Or for ongoing social video programs, conduct periodic brand lift studies—at least one per year and up to one per quarter.
For performance marketing campaigns, Google Analytics 4 (GA4) offers a new metric called YouTube Engaged‐View Conversion (EVC) events. An EVC event indicates that a user has watched a YouTube video for at least 10 seconds and has completed a conversion event on your website or app within 3 days of viewing the video.
This new metric is a game changer. Why? Because 70 percent of YouTube viewers say that they bought a brand as a result of finding the brand on YouTube. But YouTube video viewers have strong intent to continue watching the content that they’re in the middle of watching. As a result, they’re very likely to stay on the platform when they encounter an ad as part of their viewing experience.
Measuring Conversation, Amplification, and Applause Rates
MATT:
Absolutely. I’m really excited about that. I think that’s going to be great.
And there are other social media metrics that we should use as KPIs. They’ve been hidden in plain sight since October 2011, when Avinash Kaushik first defined them in a post on his Occam’s Razor blog entitled, “Best Social Media Metrics: Conversation, Amplification, Applause, Economic Value.”
The definitions have evolved over the years, but here’s what each one measures and how it’s calculated:
- Conversation rate: Indicates how many Comments/Replies on average each of your posts has received (e.g., Conversation rate = # of Comments/ # of Posts)
- Amplification rate: Indicates how many times on average each of your posts was Shared/Retweeted (e.g., Amplification rate = # of Shares/ # of Posts)
- Applause rate: Indicates how many Likes/Likes each of your posts has received on average (e.g., Applause rate = # of Likes/ # of Posts)
- Economic value: Shows how much economic value each visit from a social network brings to your site by completing your site’s goals (e.g., Economic value = Per session Goal value, which is calculated by Google Analytics when you assign a monetary amount to each completed activity, called a conversion, that contributes to the success of your business)
You and I have both been training marketers to use these metrics as KPIs since our good friend Avinash came up with them. And I absolutely love that he came up with three different metrics for measuring the real active engagement of users with your social media pages.
GREG:
Avinash is the Digital Marketing Evangelist at Google, so you’d assume that the vast majority of marketers would be using these “best social media metrics” by now. But they aren’t.
Why? Because they make most social media marketing activities look like a total waste of time and money. So, most marketers stick with vanity metrics and hope that no one notices.
But it turns out that we need to adopt Conversation, Amplification, and Applause rates as KPIs. Why? Because the social media platforms are using them.
For example, Facebook freaked everybody out in January of 2018 when they announced what the industry called the Facebook feed apocalypse. That’s when they acknowledged that their engagement metrics were a little whacked. They needed to readjust them.
What they did was decrease the impact of Reactions in their algorithm, increase the impact of Comments, and maintain the impact of Shares. So, even Facebook was using these metrics to make adjustments in their algorithm. And you should also focus on them if you want to create content that gets higher engagement and interaction rates going forward.
MATT:
And if your posts aren’t engaging your followers, then you shouldn’t expect your social media accounts to generate much economic value.
Comparing the Economic Value of All Your Social Media Accounts
GREG:
And economic value is already an important metric. Increasingly, it will become your most important KPI. Social media marketers are going to have to struggle with this ugly transition between now and then.
But guess what? You’re struggling in the right area.
After Willie Sutton, a famous bank robber from the 1930s, was arrested, a reporter asked, “Willie, why do you rob banks?” And he said, “Because that’s where the money is.”
So, if you’re not measuring economic value, then learn how to. Because that’s where the money is. And one of the things that may require you to do is connect the dots.
As Matt said eloquently at the beginning of this podcast, you need to think beyond your social media platforms. If people need to come to your website before they can buy your product, generate a lead for your sales force, or do something that can be monetized, then figure out how to create a conversion event in GA4. And, then:
- Link your property to Google Ads to make YouTube Web EVCs available in your GA4 reports.
- Activate Google signals to see conversions from users who are signed in to their Google accounts.
- Add a monetary value to each conversion event.
Finally, use Predictive Audiences to target “likely seven‐day purchasers” or create a remarketing campaign to re‐engage users based on their behavior on your site.
Now, this won’t be easy to learn at first. But it will eventually enable you to compare the economic value of all your social media accounts. So, when the often predicted social commerce trend finally takes off, you’ll be in a strong position to measure what matters.
Leave a Reply