How To A/B Test Your Meta Ads Creatives (+ Free Cheat Sheet)

[@abderrahimakouz3062]

First like and first comment from Morocco 🎉

[@VertexMarketingAgency]

Thanks :)

[@Rhuanvlogss]

Hi Cedric, great video! Do you know what I can do if I get a "below average" quality ranking for my ad videos?

[@VertexMarketingAgency]

Is it still getting the results you want? Ive had high converting ads (sales) with low quality ranking, wouldnt worry about it unless it's not getting rsults.

[@ylize-1313]

Hi Cedric, I launched an A/B test a few days ago to compare two different promotions and determine which one achieves a lower cost per result. However, I’ve noticed that both campaigns aren’t fully utilizing the allocated budget. Do you know what might be causing this?

[@NathanVecchiarelli]

So how come we don't just ad new creatives to an existing campaign? I don't really understand that, if we have a winning ad from the creative testing and add that to the main campaign will it get any ad spend? Could we view the testing campaign as a main campaign 2.0?

[@VertexMarketingAgency]

Because it is not a fair playing ground for new tests. all tests should have 3-4 creatives. you winning campaign should also have 3-4. if there are already ads in the campaign getting spend, the new creative tests will not get spend. its also a much better organization structure to have 1 testing campaign and one winning campaign for long term ads management. and you may run more than 1 test at a time (2 ad sets, 1 campaign) which would make way too many inside one campaign! Hope this helps :)

[@shafkattipu7504]

I have setup server side tracking for meta with ipv6 update sending from server side. But my problem is i am getting a dignostic in my events manager, they saying my client ip address is not matching what I am sending from server. What the problem is?

[@jaekodesigns5456]

Hi there, I used to do this but Meta keeps telling me that it causes audience fragmentation. Does this happen to you too? If so, do you just ignore it? In my case, whenever it does that, suddenly the results will become inconsistent after a week of receiving that notification from Meta. Thanks in advance!

[@zamamkhizemedia]

Thank you so much for this video you have gained a subscriber

[@VertexMarketingAgency]

Thanks for the sub!

[@Straightwaller]

So do I make 3 Ad sets for each of my 3 creatives? I like your theory and explanations but its lacking the actual set up in the campaign manager

[@codyromness]

What about testing 2 different offers - a free trial via a paid package? Should I show the different offers to same audience? Or should I split test so that audience A doesn’t see audience B’s offer

[@priyakrishnan7083]

Does A/B test variations show only one version to person. I am trying to run a study where I want to make sure one person sees one variation only.

[@ashkan-vahdani]

This is what I think the A/B testing method from meta will do. Otherwise it may show both ads to one person after time.

[@anas10ify]

Thank you so much for the video. I have a question regarding the budget, I didn't get understand it fully. So should I put a lower budget in the tested campaign?

[@VertexMarketingAgency]

Yes, I always spend less in my testing ad set since my main campaign is the one with all the winning creatives.

[@ernesttomas9358]

Why do we have to duplicate the winning ad test to the main campaign?

[@LyAds-w6e]

Hi Cedric, thanks for sharing this. Just a question on the two ad sets containing varying ads, will they have the same audience? If yes, will it not cause auction overlap?

[@theboundaryjournals609]

Do you delete old ads/adsets once you duplicate them? Curious how to keep the dashboard looking clean. Thanks!

[@VertexMarketingAgency]

No! If recommend never deleting ads. In 3 months, months, 2 years.... you want to be able to see the old data! just turn off. keep it clean by using new ad sets or campaigns

[@KENNETHJANSENMAGTABOG]

Very Cleaaaar Tutorial bro! Thank you! new subscriber here

[@VertexMarketingAgency]

Appreciate it!

[@zaidelhamdouni7724]

thanks for the video! I face a problem that Meta just pushes always one Ad, which do not give a chance for the athoer creative to get traction. Is there a slution for that other than creating a duplictae the Ad Set with the ad variation, and run A/B split test on Ad set level?

[@VertexMarketingAgency]

In any creative test you do, there most likely only going to be 1 or 2 winners. you will probbaly have to turn off the others. its all based on the algorithm, so if it doesnt get spend it probably wasnt the best one! alternatively if you really love that ad you could add it to your next text and see if it gets spend there. if it doesnt, it's a loser. thats the name of the game!

[@zaidelhamdouni7724]

thanks for the reply, i will try it out!

[@feedplan]

I am working on advertising an app, that is only available in 2-3 cities: Boston, Denver, Austin.Should I have a single ad set inside of the campaign that includes all 3 cities, or should I be creating 3 different ad sets for each city?

[@VertexMarketingAgency]

Depends on how important it is to you to know which city is performing and if you would want the control on how much spend they each get. You could test both (all together vs all seperate) and see which is better. The bigger the audience, usually the better, but can also make you have less control over where the spend goes.

[@shahzebkhan4429]

Im A/B Testing in an ad set with 3 creatives. And one performs well. When I want to test more creatives, do I put them inside that same ad set and turn off the low performance ones. Or do I create a different ad set with the winning creative duplicated into it. And then add new creative tests in there? Please explain this A/B testing in detail...?

[@VertexMarketingAgency]

Create new ad sets for each creative test!

[@sologrind9895]

Hi! This video is so packed with details! I dunno if you've mentioned it or not, but, on what level is your budget allocated on your main campaign again?

[@VertexMarketingAgency]

I like to keep the main campaign in CBO but it doesn’t matter if you only have 1 ad set.

[@sologrind9895]

@@VertexMarketingAgency sweet!

[@svv7741]

Can you also do this with testing first day dropship products? Because i wanna test this strategy

[@AlexxxxAlpha]

You just earned my like and sub. Thank you for this tutorial

[@VertexMarketingAgency]

Glad it helped

[@rosariomira9821]

Hi Cedric, thank you so much for the awesome video, super clear! 🙌 I have a question: You talk about letting the winner ad run in the test campaign instead of copying it to the main campaign until it works. We had the problem of having a high performing ad running in the test campaign and when we copy it to the main campaign it didn't work at all. So I re-activated in the test version again and it's still running well. But this creates a huge mess in the account 🙈 Do you have any recommendation/structure to differentiate tests from campaign ads? Thanks a lot!

[@VertexMarketingAgency]

You can have multiple testing ad sets running at the same time. if they are performing, don't turn them off, even If you copy them over to your main campaign! make sure you have at least 3-4 ads that have proven winners before you copy them over to your main campaign. if you make 1 test, for 1 ad that won, and copied over and only have 1 ad in your main campaign... of course it will not work well. Hope that helps!

[@rosariomira9821]

@@VertexMarketingAgency thank you for he reply! To make sure I have the correct understanding, when you say: - "don't turn them off If you copy them over to your main campaign!" do you mean let them run in the main campaign + test campaign? - "3-4 ads that have proven winners before you copy them": I should only copy testing ads to the main campaign when I have a least 3-4 winners in the main campaign? Can you elaborate on the logic behind this, please? Thanks a lot for your help!

[@PavloUlytskyi]

Cedric, hello! When you use Meta A/B testing, Meta split audiences, in your approach it does not happen. If you use ABO and 2-6 ads inside you let Meta algorithm to decide what creative is best, some of your creatives could not be delivered or have low reach. You can create A/B test in Experiments section. The ad sets will run normally outside of the test.

[@ashkan-vahdani]

Great video. My only concern is that both ads will be shown to one person after time. I thought that meta's AB testing solve this by splitting the audience for you.

[@DanielJørgensen-c6z]

Hi. Thank you very much for this content 🙏🙏 Why do you not just make the different ad variations in one ad group instead of making different ad groups? And how many ads can you put into one ad group - Only one? Thank you

[@VertexMarketingAgency]

Make a new ad set for new creative tests. You can have 2-6 ads per ad set that are active. I explain it all in here :) vertex-marketing.notion.site/A-B-Testing-Facebook-Ads-in-Ads-Manager-2ef8a5e141ed4637a0d0883d68ed7121

[@musicgrynd]

Thanks for this Amazing video!!What is the pixel option i see there?Is that something we have to install or is it just an option that comes with Business manager?

[@VertexMarketingAgency]

Yes, you definitely want to install the pixel on your website if you’re sending traffic to your website or landing page.

[@musicgrynd]

@@VertexMarketingAgency Thank you very much!!

[@LucidBoxFilms]

Excellent Video. Thank you for this! One question, would it be better to copy the winning test adset, instead of copying the ad into main adset? Since meta mentions that adding an ad into an existing adset will reset the algorithm. Thanks!

[@VertexMarketingAgency]

That’s why you only copy the ad when it’s performing better than the ads in your main campaign. So you’re re-entering the learning phase but because the ads have already been tested you know it worth it!

[@AIAffiliator88]

Whatever you do dont just edit the ad, that will skew the data

[@AmiraIbadillah]

thank you!