1. This website uses cookies to improve service and provide a tailored user experience. By using this site, you agree to this use. See our Cookie Policy.
    Dismiss Notice

How is my CBO performing worse than my winning adsets?

Discussion in 'Dropshipping & Wholesale Hookups' started by Social Panel, Dec 1, 2019.

  1. Social Panel

    Social Panel Junior Member

    Joined:
    Sep 3, 2017
    Messages:
    145
    Likes Received:
    25
    Gender:
    Male
    Home Page:
    Quick question.

    It's my first time trying out CBOs so i don't know much of what to expect.

    I had 3 profitable adsets from a previous low budget testing campaign. (CPC ranging from $0.30-$0.60)

    Once I duplicated them into a CBO with $70 budget with min spend of $11.66 across 3 adsets ($70 is my fb ads limit, tried contacting them to raise it and am waiting for response) the campaign's CPC skyrocketed to $1+. It has been a full day now and it's still the same. Each adset has a $1+ CPC.

    Did i do something wrong? Is it normal for CBOs to take longer to optimize fully, or maybe CBOs tend to be more inconsistent?

    I feel like i messed something up because i can't see how if the adsets were winning before, how they could have gotten more than twice as bad just in a new campaign. Any advice would be helpful.
     
  2. Roids

    Roids Newbie

    Joined:
    Nov 11, 2019
    Messages:
    2
    Likes Received:
    1
    Gender:
    Male
    First off, let’s remember that CBO’s are still in the machine learning process. They’re getting better every day, but by no means are they perfect or going to work every time. Definitely have their quirks.

    That being said CBO’s work best for me with 5-7 ad sets.

    Also it is without a doubt in your best interest to horizontally scale first. I don’t normally touch CBO’s until after starting LLA’s. Optimization won’t be an issue at that point
     
    • Thanks Thanks x 1
  3. Social Panel

    Social Panel Junior Member

    Joined:
    Sep 3, 2017
    Messages:
    145
    Likes Received:
    25
    Gender:
    Male
    Home Page:
    Thanks for the tips!

    I had tried LLAs as well but they were unprofitable, is there a recommended amount of videos views you would suggest before trying? I have about 1600 75% and 900 95%
     
  4. CBO can still be a hit or miss, it takes 3-5 days to fully optimize and it depends on the account you have as well - something that no one wants to talk about :)

    When everything that makes sense fails this is how I test things -> Create a CBO campaign with your tested/best-performing audience and duplicate it 3-5times (depends on how big your budget is), and start running ads. Monitor your ad spend and start cutting once the ad set spends $10, rinse and repeat. This approach sounds ridiculous but always made a ton of money :) Don't forget to use your best performing ad as well, where you have enough social proof.

    Regarding LAA, CBO should work pretty well if you have enough data. What I usually do is combine so-called 'super LAA' together in one ad set --> LAA 1% buyers, ATC, IC, highest LTV, top 25% spent on-site, etc.
     
  5. Extruser

    Extruser Junior Member

    Joined:
    Dec 2, 2017
    Messages:
    188
    Likes Received:
    124
    Gender:
    Male
    Real question. Not trying to be an ass here.
    But do you really know how CBO works?
    I mean, it needs more data. $70 budget is not enough.
    You would need at least 3 ad-sets with 3 ads each and
    spend $20 on each ad during 5 to 7 days. So $900 - $1260 budget.
    And that's a minimum, since you cannot laser-target at that point,
    unless you know your audience really well.

    Happy to hear input from others.
     
  6. Social Panel

    Social Panel Junior Member

    Joined:
    Sep 3, 2017
    Messages:
    145
    Likes Received:
    25
    Gender:
    Male
    Home Page:
    Thanks for the tips. Yeah i had someone else tell me a similar thing, to duplicate the exact same ad sets multiple times into the same cbo and for whatever reason some adsets will work a lot better than others. I assume that has a similar affect when duplicating a cbo multiple times. I've setup a new strategy (not cbo yet) of 10 adsets single interest with tons of different creatives each (carousels, different videos, different thumbnails, pictures, first 3seconds etc.) I had also put some of the adsets as LLAS. I Have 1600 75% VV, is it normal for that LLA adset to sometimes perform worse than my single interest broad targeted adsets? Or do i not have enough data yet? I have a lot more 25-50% but i feel like that just wouldn't convert well because at that point they really haven't watched much of the video.

    Haha i guess i didn't. I thought it would use automatically spend budget on the better performing adsets and optimize them automatically based on the data coming in. I guess for now I will stick to regular campaigns until facebook lifts my limits.
     
  7. Every time you duplicate an ad set you 'hit' different pool of audience (with a small overlap). That's why some work and some don't.

    Testing a ton of creatives make sense when your budget is huge. I never ever test more than 3 (max 5) at the time and optimize the best performing. If Facebook pushes the budget to one ad only, and data shows that different ad has potential I shut other ads to see how it performs. Keep in mind - the more creatives you have, the longer it takes to collect enough data to prove the winners.

    If your LAA is not performing well, you need more data. One thing you can do at the beginning and improve them is to narrow down the targeting with broad interest.

    Good luck!
     
    • Thanks Thanks x 1