Facebook recently started showing advertisers a new and pretty awesome metric: the Relevance Score.
The new metric is a score on a scale of 1 to 10 that rates an ad based on how relevant it is to it’s target audience.
What is the Relevance Score and why does it matter?
The relevance score is a magic number. It’s a score from 1 to 10 that, at first sight, will let you know how relevant an ad is to its audience.
Note the wording. Relevance, not quality.
The scope of this number is not to tell you how good your picture or your copy is. The main goal here is to make you understand if your ads’ demographic target is considering your message relevant to them and thus engaging with it.
An ad made by the best designer in the word, with the best copy ever written will still have an insufficient score if it’s promoting women’s clothing and it’s being targeted to male teenagers 13-18 years old.
On the other side, an ad with a decent design and copy targeted to the perfect audience, will enjoy a healthy relevance score.
Here’s an example from Facebook:
You may be wondering, “Why does this matter? It is just a number.” It is much more than just a number. The relevance score is the synthesis of many metrics: Engagement, Clicks, Conversions, Click-Through Rate, Negative Reviews. With just one number you can quickly understand if an ad will be a winner or a loser.
The Relevance score has also a direct link with your advertising cost. When Facebook has to decide which ad to display to a specific user, they’ll always prefer to display an ad which they consider relevant to that user. If yours is not, it’ll quickly become very expensive to advertise.
Don’t believe me? Just Watch.
This Ad we’ve been running had a poor targeting and costed us an average of $0.142 per website click. The Relevance Score is 2.9
This is the same exact ad but targeted to a Website Custom Audience made out of users who visited our website in the last 90 days. The Relevance Score is 8 which leads to a cost per website click of $0.03!
With the same budget we generate 4 times more clicks. 1,103 Clicks to this blog post compared to the 278 of the other campaign. Shares, Likes and Page Likes are through the roof as well!
This is just one example of how important the relevance score is.
How to check the relevance score
It’s extremely simple. Both in the Ads Manager and AdEspresso.
Simply access the Ads Manager select your campaign, then an AdSet you want to analyze and in the list of Ads you’ll be able to see Facebook’s relevance score for each one. Clicking on the Ad and then on the “Relevance Score” Tab will show more details:
There you go. You can also see the volumes of Positive and Negative Feedback.
Checking your Relevance score in AdEspresso is even simpler than that! As soon as you access your campaign’s dashboard you’ll be able to see the trends of your campaign’s relevance score and engagement over time.
You can of course also see the relevance score for each ad in your campaign by clicking “All Ads” in the campaign’s menu.
Since we display in the same page both AdSets and Ads, you’ll also be able to see a bird-eye view of the average relevance score of each AdSet.
Data Time! What Relevance Score means for you and what influences it
Now that you have a good idea of what the Facebook Ad relevance score is, let us check some data.
We’ve analyzed a subset of our database: 104,256 ads created through AdEspresso in the last 45 days. The time range is due to the introduction of the Relevance score. Earlier than 45 days ago, very few ads had a score at all.
First we checked the distribution. I would have expected to see a classic bell curve with most of the ads scoring between 4 and 5. Turns out, our customers are amazing and 63% of the ads had a score of 6 or more. 8 was slightly the most common relevance score.
It’s great to see most of the ads are targeting a relevant audience.
I’ve already mentioned that your ads’ relevance will impact Facebook’s willingness to display it in users’ newsfeed and consequently, how much it will cost. Let’s see some numbers to prove that:
|Score||Ads||CPC||CTR||Avg Shares||Avg Likes||Avg Comments||Avg. Website Clicks|
Exactly what we expected. The higher the relevance score is, the lower the Cost Per Click is and the better the Click Through Rate. The difference is really big if you reach a relevance score of 10- the CPC is extremely low and those ads generated a huge amount of clicks.
The next question is: Are the CTR and the CPC so good because of the high Relevance Score or vice-versa?
I tend to towards the latter. The relevance score is not a standalone metric like the number of clicks or impressions, it is a calculated one. The relevance score cannot affect the clicks that an ad will receive, It’s exactly the opposite. If a Facebook ad has a great design and is targeted to the right audience, it’ll have a very good CTR and engagement. This will lead to a lower CPC and to a higher Relevance Score.
The relevance score is far from useless, and to a certain extent you can avoid checking multiple metrics and simply focus on the Relevance Score to understand which ads are performing the best.
Finally, let me do some speculations and try to guess the metrics that more influence the Relevance Score:
- Negative Signals: When a user hides your ad from their newsfeed, that is a strong indicator that you’re not targeting the right audience.
- Campaign Objective: I think this is the strongest positive influence you can get. If your campaign is aiming at likes and you receive a lot of likes, that’s a good sign. The same goes with Conversions, Website Clicks and so on. The better you perform on your campaign’s goal, the higher your score will get.
- CTR: The higher your click through rate is, the more relevant your ad is considered by Facebook. This is common sense, as no one would click an ad that is irrelevant to them.
- Shares: My theory is that Facebook believes the sharing an ad is the strongest endorsement on its quality. The numbers above seems to back this theory with a pretty strong correlation between the shares and the score.
- Likes & Comments: While the data doesn’t show a clear correlation, I think likes and comments are considered by Facebook as positive quality signals. Comments may be a bit trickier as without understanding the context is tough to attribute them a positive or negative value.
How do you improve Your Facebook ads’ Relevance Score?
Remember what this is all about: Relevance.
To increase your Relevance score, you simply have to craft appealing Ads that directly address the needs and wants of your very specific audience.
Here are four actionable tips to improve Facebook Ad Relevance Score:
- Use Buyer Personas: Don’t create ads for just anyone. Identify specific buyer personas and address their problems and needs in your ad. This was our first point in the recent guide on Facebook Ads Design.
- Split Test Everything: Before scaling your campaign, test it. Test multiple designs and multiple audiences to find the combination with the highest performances and Relevance Score.
- Don’t get over-obsessed: In the end you are advertising to make money. As long as your Ads have a positive ROI, you are doing well. A low relevance score means you can do even better, not that you should stop advertising. I personally have campaigns with a score of 3 that still generate a lot of money for our business.
- Try Interests Intersection: Instead of targeting very broad interests, try to use the interests intersection feature to laser-focus your targeting. Promoting your gluten free, vegetarian recipes? Targeting only people who are only interested in both Gluten Free AND Vegetarian rather than Facebook’s default OR logic could double your performances.
- Monitor the Frequency: A high frequency can kill even the best campaigns. If your frequency is too high, refresh your creatives or change the targeting.
Your turn now!
I hope this data backed analysis of Facebook’s Relevance Score helped you better understand how it works and how you can benefit from it.
Now it’s your turn to go check your campaigns and improve them.
While you are at it, why not share your findings in the comments below? What is your average Relevance Score, what worked well to improve it?
Let me know, and we’ll compare findings!