One Facebook ad's reach is way higher than the others. Why?

6 replies
Hi guys!
Wondering if anyone has insight into the Facebook backend as to why nearly identical ads are performing dramatically different.

Here's what i did
1. Created a campaing
2. created an ad set
3. created an ad with 4 different pictures and selected single picture ad so it created 4 unique ads each with a different picture.

My reason was to test the effectiveness of the various images. What confused me and also made me post here is the amount of variance in the reach. I was hoping to test the click through and compare the stats, but after seeing one of the four ads get delivered way more i was concerned Facebook might have some type of algorithm that effect delivery depending on the image

The only reason i can think that one ad is ads reach is way higher than the others is a Facebook bug or feature.

Any thoughts?

FB Reach Stats
Ad 1: 1
Ad 2: 6
Ad 3: 5
Ad 4: 467
#delivering #facebook #set
Avatar of Unregistered
  • Profile picture of the author Tsnyder
    Is the high performing ad getting significantly more engagement?

    I've had that happen. When you start generating engagement
    FB serves that ad more because it appears the audience is more
    interested in the content. The more engagement you get the more
    FB serves that ad.
    If you knew what I know you'd be doing what I do...
    {{ DiscussionBoard.errors[11191455].message }}
    • Profile picture of the author Destinationsound
      Hmmm. Not off the bat but good point. It could have been a snowball effect. It might have gotten an early spike which caused Facebook to push it harder.

      Another possibility i was thinking is, it has the brightest most contrasting image. Curious if FB robots pick up on that and see it as a more clear and vibrant image, therefore feature it.

      AI is only so smart I guess
      {{ DiscussionBoard.errors[11191458].message }}
  • Profile picture of the author AdHustlers
    Whoa that is a dramatic difference! How long have the ads been running? I noticed in Facebook's "A/B testing best practices guide", it says tests should ideally run for 3-14 days ( Perhaps letting it run longer will even out the reach?
    {{ DiscussionBoard.errors[11192202].message }}
  • Profile picture of the author sprinthurdles
    I've noticed FB does this when you run more than 1 ad in the same adset.
    For this reason I only run 1 ad per adset. You can still split test the images, but in order to get good reach to all 4 ads it'd be best run seperate adsets to all 4 ads.
    {{ DiscussionBoard.errors[11192291].message }}
    • Profile picture of the author Destinationsound
      I turned off the other 3 because they weren't performing and I didn't want to waste any funds on them.

      Great idea about various ad sets. I'll try this next.
      {{ DiscussionBoard.errors[11192309].message }}
  • Profile picture of the author peterparker1997
    When I see the add in my Screen, It is not comfortable to me
    {{ DiscussionBoard.errors[11192351].message }}
Avatar of Unregistered

Trending Topics