Terms like “Relevance Score” and “Quality Score” can seem vague and mysterious to people who just want to put some money in the PPC slot machine and have sales come out. Facebook introduced its Ad Relevance Score in 2015, but many advertisers still struggle to understand it, or how to fix it if it’s struggling.
If you’re one of those advertisers, read on. This is everything you need to know about Facebook Ad Relevance Score.
What Is the Facebook Ad Relevance Score?
With the explosion of Facebook Ads and the News Feeds getting ever more crowded, it made sense for the social network to create an ad quality metric. This also added a layer of complexity. Advertisers had another “thing” to figure out if ads were under-serving or costs got really high.
Your Facebook Ad Relevance Score is a rating of 1-10 after it has at least 500 impressions (yes, that’s pretty quick). The score is calculated daily based on, as Facebook says, “positive and negative feedback we expect from people seeing it, based on how the ad is performing.”
Facebook goes on to define “positive” as things like shares, likes, or other actions that help you achieve your objective. Meaning that yes, the criteria for your Relevance Score can change a bit depending on whether you’re running a campaign with an objective of video views vs. one for link clicks.
“Negative” feedback is anything like when people hide your ads. Though Facebook doesn’t explicitly say so, it’s also safe to assume that anything not meeting your objective (i.e., people not clicking, etc.) also contributes to negative feedback.
This isn’t surprising. After all, Facebook is a social network. Facebook rewards you for generating more interaction and interest — it’s their value proposition and they have to protect it (in the same way AdWords protects the quality of search results by having a Quality Score).
Does Relevance Score Really Have an Impact?
I was skeptical of Facebook’s Ad Relevance Score when it came out. A lot of AdWords advertisers didn’t consistently see more favorable CPCs as we improved Quality Score, and the word itself became another dreaded metric that can distract from overall account goals.
However, it’s pretty easy to see with Facebook that when you’re able to improve your ad relevance you also reduce your costs
I decided to specifically test this in an account that was doing well socially already.
To understand this method, it’s important to understand one nuance first.
Ad IDs & Sharing Them
When you create an ad on Facebook, it automatically generates an Ad ID.
Copying and pasting that ad into another Ad Set creates a new ID. Even thought it’s the same ad, Facebook will treat them as being different and it won’t retain the Ad ID.
Social interaction on an ad is tied to the ID level. What that ultimately means is each ad unit will keep the social interaction to just that ID — it doesn’t share it with the otherwise identical ads because the ID is different. Each ID also has its own Relevance Score:
You can get around this by creating an ad and then pasting its ID into the option for “Use Existing Post” when you create a new ad. This will share that ID, and all the social proof it accumulates will display on that ad for every ad set that it’s used within:
Important: As with any ad, if you update the copy, link, or anything like that, it will reset all of that social proof. This carries even more weight once you share an ID because it will affect multiple ad sets with the stroke of a key.
Pushing That Relevance Score
I decided to test what happened with a focus on Relevance Score. This meant we wanted to try and house all the social proof on one ad ID in order to maximize impact. Otherwise, we’d have a bunch of disparate ads with each having their own social proof.
I took a well-performing ad that had duplicate versions with different IDs running in different ad sets. The average Relevance Score of them all was around 8. The social proof on them all was similar and their Relevance Scores were the same, so we just picked one at random.
I took that ID and pasted it into the other ad sets. This would ensure that moving forward, any new social proof and any change in Relevance Score would be concentrated in this one unit. It would amass the social proof faster because it wasn’t being distributed among disparate Ad IDs.
The shared post ID started to run, and then I did one more thing: I added a Page Post Engagement campaign (now just called “Engagement” objective type), and for the creative I used the same Ad ID. (I threw a couple hundred bucks towards it only because the population size warranted it, but the same methodology can be used with just a few bucks a day.)
This means that while that ID was running in the ad sets focused on converting to sales, it was also running in the PPE campaign, accruing social results simultaneously. Within a few days, the Relevance Score for the shared ID was hitting a 10.
I pulled the CTR, CPC, and CPA data for the ads when they ran in different groups (called Non-PPE version) vs. when I previously ran the single ID with the extra dough put towards accruing some social proof (called the PPE version). Other than as outlined above, nothing else changed, including the targeting.
It appears the additional social proof gave us a leg up here.
Again, the additional social proof seemed to help here. (I didn’t analyze it at the time, but it would have been interesting to see if the offset in CPC saved enough money that it paid for the social proof.)
Finally, the ultimate number that matters:
Wow. That’s a huge difference, and definitely statistically significant with the audience size we had.
I’ve run this test in a few other accounts, and the results were similar. Sometimes the result wasn’t as marked, but it was still usually there.
Clickable, Shareable, Likable
What I liked best about running this test was that it helped take the hypothesis of Relevance Score, and proved it. It’s easy to say “create great content,” but it’s another thing when you see how it directly impacts the bottom line of what you’re selling. It’s worth the extra effort to test that video, or do something more creative than just throwing a stock image up there and hoping for the best.
Like every ad creative on Facebook, eventually it ran its course and it was time to swap out.
The beauty of this method, however, is it allows you to launch a new creative and secure a higher relevancy score faster. It helps alleviate the peaks and valleys that can be associated with launching new creatives, and get relevancy score working in your favor, faster!
In-post images: Screenshots by author. Taken May 2017.
Subscribe to SEJ
Get our daily newsletter from SEJ's Founder Loren Baker about the latest news in the industry!