How to Test Facebook Ads - Best Facebook Ad Testing Strategy for 2021

[@lymphomasymtomps]

One of the best videos about FB Ads

[@Vendiendoyviajando]

Why you have 809 views... this is amazing!

[@AndrewHubbard1]

Thanks!

[@kenettelouis3568]

Great video - I am officially following you now. Question: what size audience is best for running the ads?

[@AdA-rl4eo]

Thanks for the video, always great content 👏🔥 ps. Anyone ever told you that you look like Cooper Cronk?

[@AndrewHubbard1]

haha I haven't heard that one before!

[@cjmarchi]

hi, great content, thank you! What do you think of testing all those interests matched with engaged buyers?

[@localbusinesschampions9502]

awesome video. straight to the point. no fluff. super helpful thank you. keep it up! this channel deserves wary more than 4k subs.

[@AndrewHubbard1]

Thanks!

[@Ctobull]

Hi Andrew great video! My questions is if my ads have not made any sales after 3-4 days should I keep them running?

[@dimonteex]

What are we advertising/offering in the adverts?

[@thisorthat5790]

What about overlapping of audience issue? Dublicating adsets for testing in same successful audience again and again .

[@alishahzad2869]

This guy is so amazing n so to the point, I actually appreciate him!!! 👍

[@AndrewHubbard1]

Thanks Ali! Appreciate you

[@ajdaam5898]

What about learning limited ? What about adding budget to working ad sets? Should I just create CBO and put all winning ad sets there? and keep testing new ads and interests in my original ABO campaign ?

[@bekc1626]

Thanks heaps for the vid! Quick Q...to anyone seeing this. If my 3 Audiences/Ad sets have an individual ad in each one that has hit my KPIs really well (with minimal sales albeit), but 1 audience/ad set is a bit better than the other 2, should I actually just turn off individual ads in those ad sets and replace with new ads, or do I turn off the 2 slightly less performing ad Sets and dedicate my budget to the 1 ad set in order to test creatives?

[@AustraliaAupair]

Thanks for the content! This helps a lot. Are there any other ways to do testing? For me does it seem like this would be the only solution and it sounds so logic that I wonder if everyone is doing it like this or how are others are doing it?

[@tonybarton5152]

Hey Andrew, good to hear and Aussie voice, also I suspect relevance to local biz. Any ideas re real estate fb ads ? Regards Tony

[@getslimnogym493]

Hey nice vid! Just got a quick question. When you group multiple interests in one ad set, how do you know which interest is bringing results? Why is the one interest per ad set outdated?

[@naveedtayyab193]

Same question from my side. Sir answer would be appreciated.

[@YoLo-dt7ox]

I was just gonna ask the same.. I see lots of advertisers using the one interest per adset strategy to test interests. There is no other way to find out which interests are working or not right? What I think tho is that you could group smaller, unpopulair or hidden interests together in one adset to get atleast the minimum reach for Facebook to find the right people. Because testing one small interest at the time wouldn't work. In that way you don't have to compete on 1 big populair interest that everybody else is targeting too. Does that makes sense? I'm just thinking out loud here:P

[@MsDrwalker]

super helpful! Thank you

[@Freedomskills]

Powerful Insight ad set testing really helped me out

[@AndrewHubbard1]

Great to hear!

[@Freedomskills]

Hi Andrew £1000 ad budget how many ad sets would you recommend testing?

[@akid0305]

Dear Andrew, you are one of my favourite "teachers" for digital marketing on You tube , amazing videos you have. And I have few questions. We are running some FB ads for a mobile phone operator. First one is an awareness campaign and I would create lookalike or custom audience and the other one is a conversion campaign (conversion is buying a post-paid or pre-paid SIM on our web site) where I would create no audience since we have the FB pixel. Do you think that’s a right approach.

[@silviyad9621]

Thank you so much. Your videos have the best content !

[@crystalthemarketer6546]

For the ad set budget for audiences is that $25-$50 per day or lifetime spend over 5-7 days?

[@Thibaut.daviaud]

Would having the same audience in different ad sets mean a single customer could see all the ads or would it be smart enough to not show it to the same people due to the fact that it's part of the same campaign?

[@s-carrshymahdi9003]

“Target look alike and target interest” can you elaborate?

[@mohammedrafeequem4224]

thanks for great information

[@AndrewHubbard1]

You're welcome

[@alo5323]

Hi Andrew thank you for the amazing video . Quick question , I’am a little confused about what you said about creative ad testing in multiple ad set from time stamp 10. 28 to 10.46 . For example if I have three ad set (audience A, B,C )each ad set with 3 similar ads ( creatives 1,2 &3) Audience testing suppose A is the winner. Creative suppose 1 is the aggregated winner Now next step if I want to test additional 3 more creatives Duplicate the winner ad set A containing winner creative 1 remove previous 2& 3 and include additional testing creative suppose 4,5,6. Pause the original Ad set A along with all other poor performing ad set and only keep the duplication running Is that what you are saying ? Sorry for the lengthy query.. tried squeezing it as much as possible.

[@marketingaleco7812]

If you test audiences, is it important that the size of the audiences are (almost) the same? Like now I've imported a customer list but the size isn't available. Thanks in advance.

[@gpmikegeorge]

Also, is this the same way you test for warm and retargeting campaigns. How do you break down your budget for the entire funnel, to include the testing? Thank you so much!

[@akmalhakimhamid1807]

Hey Andrew! Great video. May I know what's the reason behind duplicating an ad set for the next set of 3-4 creatives? Why can't we test all our 9 creatives under 1 ad set?

[@AndrewHubbard1]

Good question! If you have more than that many creatives in an ad set, Facebook tends not to allocated much spend at all to some of them

[@akmalhakimhamid1807]

@@AndrewHubbard1 thanks Andrew!

[@christianpicardo2683]

Hi andrew may i ask how facebook selects the winner of the bidder of selected interest? For example advertiser A selects Dogs as interest with budget of $100 testing budget, then advertiser B selects dog ,then narrow it to bags with$50 budget , do advertiser B still gets the audience interest or wins the bid on interest dog+bag audience? Since he has different audience to advertiser A? Other is in testing , is it ok to use CBO then put it on a bid cap? They say it will reduce the cost of cost per message(im running a messaging campaign) number 3 question is ,what is the ideal number of purchases before i can create a custom audience snd look alike audience ? Im only using facebookpage and running messaging campaign i dont have a website

[@evanhares8548]

This video really helps. Just to clarify, you're saying test the ad sets first with one ad each and then pick the winner and then tests different ads? Or test multiple ad sets with multiple ads all at once?

[@kenettelouis3568]

What is the answer to this question?

[@kenettelouis3568]

Set up 3-4 ads inside the ad set

[@franciscoisidroi2646]

Hi Andrew! Thanks for this very insightful video! Just curious, if we are to duplicate the winning ad set everytime we go to another phase of testing, does it mean we will turn OFF the original ad set everytime we test new elements? Thank you.

[@sagarthacker2339]

Franscisco, to simply answer your question, you should not turn off your original winning adet when you're about to duplicate for the next test phase because, as long as it is producing results under your KPIs, there's no point in turning them off. Hope that answers...

[@AntonioGallo73]

I didn't quite understood how do you test different copy. From the winning audience with 3 images you pick only 2 and then create 1 adset with 4 images: A1, A2, B1, B2 ? What if I've like 5 images that I wish to test with 2 or more copies? Thank you :)

[@dennisdeheza]

This is gold

[@csabapatho7781]

Hi Andrew, awesome as always. How do you manage testing creatives/copy when you have more winning audiences or really do the testing only with the top performer-audience? Or do we create a new ad set for the top performer audience with 4 new creatives and duplicate this new ad set for the other winning audiences? For example: I have 4 winning audiences, I create a new ad set with new creatives/copy for the top #1 performer audience and duplicate this ad set for the other 3 audiences with the same ads in it? Thank you for your time and answer!

[@EvgeniTerziev]

Hey Andrew! Thank you for the great video! I'm in the process of setting up campaigns for a Christmas present we will sell in Bulgaria. It's a physical calendar with 365 challenges for every day. We have 3 variations of the calendar: 1) For couples, 2) For teens and 3) For everyone. All of those have the same target audience. When setting up my testing campaign I've currently created 4 ads - 1 for each product leading to the product page and 1 for all the products altogether leading to the home page. Is this a correct way of setting up my test? With this setup, have I substited my creative testing (titles, video, photos) with an offer testing? How would you approach an account like that? Would love your 2 cents on this ad account!

[@EvgeniTerziev]

Would love your opinion on this!

[@David-eq7nj]

Hi Andrew. Quick question, once you identify your winning audience and you move on to test the creative is it advisable to turn off the campaign for Audience testing so you don't waste budget?

[@harshaagarwal7572]

Hi Andrew. The video was very insightful. You have used conversion objective to test. Is it a good practice to use objectives like awareness will testing to reduce the cost?

[@akmalhakimhamid1807]

Hi Andrew! Another question. Let's say I have a few different services I'd like to advertise. How should I go about it? Should I create a separate campaign with the same objective? Or a separate ad set of the same audience but within the same campaign? Or simply add more ad creatives within the same ad set and compete against different creatives that sell different services? For example, I want to advertise for scuba diving training & swimming & dive/pool gears. How should I go about this? Thanks in advance!

[@healthyconscious3883]

Great explanation! Keep making content

[@AndrewHubbard1]

Cheers!

[@centralfreedecentlife2522]

Thank you so much for imparting valuable knowledge to the general benefit. #AndrewHubbard

[@AndrewHubbard1]

You bet!

[@CTtheDisrupter]

Great Content! Kudos!

[@clubagency8406]

Hey Andrew, thanks for making this video - a couple things to note - I ran this strategy to the T, however the 3 ads in the ad set are not spending equally - the ad set is though but not the ads are not - how do we make the ads spend equally without FB determining what to optimize

[@lollots82]

If that is what you want to do you would have to selectively switch on/off some of the ads to allow them all to spend a similar budget before making your call. However to an extent you also have to trust the platform - if it picked one ad over the others there is likely to be one performance-related reason and pushing more budget to the other ads shouldn't improve your results

[@izakniemann5734]

Thank you for this video. However, you confuse me a bit. As far as I understand, on top is the Ad Campaign, then the Ad Sets, and underneath that the actual ads. Why do you talk about the different "Ad Sets" when in fact it's the ads you refer to? What is it that I miss or don't understand? Yes, I am new to FB marketing.

[@truongvo419]

What to do if 3 ads are not evenly distributed?

[@AndrewHubbard1]

Run the ones that didn't get distribution in a different ad set

[@truongvo419]

@@AndrewHubbard1 thank so much

[@gpmikegeorge]

Really enjoyed this video. Do you have any thoughts on minimum size audience to get meaningful results while you're testing? Thanks!

[@AndrewHubbard1]

Bigger is generally better now. As a rough guide I always aim for at least 1 million people.

[@gpmikegeorge]

@@AndrewHubbard1 Thank you! Thoughts on audience size when trying to be hyper localized? (I want to target Ottawa, open population is roughly 1 mil.) Thanks for the awesome info and taking the time to rely!

[@klajter2]

Good vid! One question: is it ok use 4 videos per ad set, but the diferent between videos is in first 3 sec.?

[@AndrewHubbard1]

Sure - just test it and see 🙂

[@marial.6160]

Andrew, do you do the testing on trafic campaigns or directly on conversion?

[@AndrewHubbard1]

Hey Maria! Always directly on conversion

[@samirdanial9346]

Thank you

[@joacole]

Hey Andrew, great content as usual Quick question: why when duplicating adsets you didn't go for an A/B test?, Is it that you don't recommend doing this? I just want to make sure that the audience is not overlapped and to avoid cannibalisation of my budget. Thanks in advance 🙌

[@AndrewHubbard1]

I've seen weird results with it personally so I stick with this method. Give it a try if you want to though! You might find you prefer it. Overlap generally isn't a problem unless you've got small audiences.

[@pascallanetzki1008]

Hey Guy, A really cool Video. I want to know what is the diffferent between an AB Test and testen 3 images or videos at one adset ?

[@EvgeniTerziev]

So, I was actually wondering about this too. What I found is that by using the A/B test feature Facebook only shows your ads to the same people from your audience making it a pure A/B test. If you put 3 ads into 1 ad set, the first ad can be seen from segment A, then your second ad from segment B, and then 3rd ad form Segment C. However, I find that people are generally not using the A/B test feature and simply just put all 3 ads into 1 ad set. Essentially if each ads gets 5K impressions this would be more than enough to conclude the test. Hope this helps.

[@ronaldbrugger4770]

thanks for the quality content

[@AndrewHubbard1]

Glad you enjoy it!

[@graphicverse4976]

Is it better to manually duplicate and test or use Facebook A/B testing feature? Also, Testing creatives, the ad set doesn't allocate budget evenly to the ads so there are two variables... what's the solution for this? Should I duplicate ad set and have only 1 ad in each or just use the Facebook A/B test feature on the Creatives? Likewise for Headlines and copy, we don't know if two of the duplicated creatives get the same budget so it isn't a fair test at all. What is the solution for this? Please reply

[@AndrewHubbard1]

I prefer to duplicate and test. Correct, it won’t allocate evenly. Usually with 2 ads in an ad set both with get enough spend for a statistically significant result, but if not just don1 per ad set.

[@creedje531]

Hey so I made a facebook business manager and i've never had a facebook business account before or have ran any previous ads. They are asking me to verify my business with business documents. I'm using shopify and im dropshipping my products. I've already sent them my license but now they want like business docs. What would you recommend i do? I don't really want to get an LLC for my shopify business. I feel like im the only one with this problem.

[@goodnews6259]

Thanks Andrew. You usually dont reply to comments I see...but... There is a reason its called CBO (campaign budget optimization) - because it optimizes for the best performing one? Why not make use of FB's intelligence and get the winners for cheaper?

[@AndrewHubbard1]

Have you tried both ways and is CBO always cheaper? Really interested to hear more about the data you’re seeing

[@florakija]

For headlines and copies do you create separate ads or do you use Facebook’s option to make various headlines/ copies in the ad itself? Wouldn’t that quicken up the process? Thank you 😊😊😊

[@AndrewHubbard1]

Separate ads. The problem with facebook’s option to add multiple is they don’t show you a breakdown of the data so you have no idea how each one is performing

[@dragannedelkovski4337]

Andrew, thank you for this amazing video. My question is when we are testing adsets and we find that an asset is a loser we pause that asset and ad a new asset that will compete against the previous winner? And if yes isn't the winning adset in lead with already getting impressions and clicks against the new one that has nothing?

[@AndrewHubbard1]

Glad you enjoyed it! Competing isn’t usually a problem with big audiences

[@jennifertiano4492]

Hi Andrew, When you are testing the creative and headlines (after audience testing) are you using CBO or still Ad set Budget Optimization? Let me know Thanks

[@AndrewHubbard1]

Hey Jennifer! Still ABO at that stage

[@FaizDigitalExpert]

Amazing ❤❤❤👏👏👏

[@AndrewHubbard1]

Cheers Faiz

[@vitaliylyubezhanin7439]

if you're duplicating the ad sets for creative testing, doesn't that mean your audience will compete against each other and increase the price of your results making it more expensive for yourself?

[@ryanwalker2340]

Hi Andrew! Been having a trouble when duplicating ad sets. When I make audience changes (my vertical age and gender is very important) it changes the settings of the original ad set I duplicated, which defeats the purpose of the duplication. Is this a new wrinkle Facebook added? Is there a workaround? There was a time, years ago, where I could easily duplicate ad sets this way and now I'm at a loss.

[@AndrewHubbard1]

Hi Ryan, that's a really strange one! I've never seen that happen before. Has to be a bug. I'd try resetting your ads manager (There's an option to do that in the top right corner (button with 3 dots). And maybe even try another browser. If that doesn't work, FB support might be your only option

[@FrankFrisk-b8s]

So, for the first round, we're testing different audiences in the ad sets, but in the individual ads we're using different creatives? Shouldn't the creatives and copy be the same if we were to only test the audience? Like different audience, same creative, same copy first round. Winner audience, different creative, same copy second round. Winner audience, winner creative, different copy last round?

[@ewingsugaringstudio]

This makes sense.

[@jerrykarczewski]

Do you need to select split test or just let the ad sets run?

[@AndrewHubbard1]

Hey Jerry - I just let them run

[@jerrykarczewski]

@@AndrewHubbard1 awesome thank you. I have the same two ads running on 3 separate ad sets rn and fb is sending out 1 ad a lot more than the other. The results of the under delivered ad have gotten much lower cost per lead. Still in the learning phase tho. Is that normal?

[@aliecommerace4864]

sir where were you all this time

[@AndrewHubbard1]

😂

[@kaoskaosdistro]

Daging ini

[@leandrovaldez5588]

I love your jokes! 😁

[@AndrewHubbard1]

Yes!! I knew there had to be someone out there 😂

[@goaway1885]

This guys trademark is huge arms leaning over position and abercrombie tshirts

[@impulse2]

Using this video as a point of reference while making my optimizations live as per your advice, I find the flow of logic is unorganised and turbulent, there is lack of logic towards a systematic approach to build my adverts out by following the video's advice step by step, the steps lack attention to detail and miss a forward approach, however overall the advice is good for advance advertisers, if you are intermediate or inexperienced I would suggest rewatching this many times and skipping to the right steps in order to find the natural flow of logic. @Andrew please take my advice as constructive, if taken correctly then it would be great if you can make a more detailed video focusing on a step-by-step process for us practical learners. Otherwise, well done on the effort and work put in, I find making videos soul-destroying, partly because I am introvert, but it's got to be done, anyhow well done on the hard work and grind that went into this, being persistent will pay itself off.

[@willydemadrid]

bro seriously i have take the time to listen the full video but people are listening to you you can't test a product for 5 days 3 days max you already know if you are making profit or not you can't spend 100 per days during 5 days just to test if you are not profitable and you repeat that five time you lose 2000$ and time 20days. more than 3 days means that you are profitable and you are making optimisation, and prepare yourself to the scaling the goal is not to sell but to make profit . if you spend 500$ in testing and if you make let us be mad make 5 sells for 50$ each you are spending 500$ to make 250 CA and in that 250$ you should remove the price of the product. your strategy for me is not good regards