Home > Paid Traffic Sources > Facebook & Instagram

Best Way to Test New Creatives (6)


08-16-2020 04:12 AM #1 rtking (Member)
Best Way to Test New Creatives

Hey guys, been running ads about 5 years now on many different platforms, but never Facebook until recently.

My question is: What is the best way to test NEW creatives on Facebook?

I have a campaign / ad set that is performing okay. It’s breaking even most days and profitable some days at $350 / day spend.

I want to increase profits more before I attempt scaling so I know I need to test more ads, but I’ve experience massive volatility when modifying Adsets. So I’m assuming leave the original adset alone?

Also, I’m running CBO on the one that’s performing well. Which I hear is bad for testing new things?

What would you guys do?


08-16-2020 04:47 PM #2 stickupkid (Senior Moderator)

Test more images/videos with same settings as you have now. On same account duplicate campaign and change images only (or dont change anything, since every campaign acts differently, also an exact duplication)

If you are already on the "bottom" of cost you probably need to improve prelander and/or offer. Without stats its hard to tell which part is causing most damage.
Profitable or potential campaigns, leave them and change nothing indeed. Better duplicate and adjust that one regarding your wishes.

For me CBO isnt cheaper most of time, but maybe my setup of camps-ad sets-ads is not how it should be to get most out of CBO.


08-16-2020 09:33 PM #3 rtking (Member)

Quote Originally Posted by stickupkid View Post
Test more images/videos with same settings as you have now. On same account duplicate campaign and change images only (or dont change anything, since every campaign acts differently, also an exact duplication)
Would it hurt to duplicate twice? Once without changing anything, and once with changing images only.

Quote Originally Posted by stickupkid View Post
If you are already on the "bottom" of cost you probably need to improve prelander and/or offer. Without stats its hard to tell which part is causing most damage.
Not sure what you would qualify as bottom of cost, but my last 7 day CPA is about $29. Payout is anywhere from $22 - $45 depending on various factors. My ads CPM is $16.83 and CPC (link clicks) is $2.62

Here is ads manager screenshot for past few days

Click image for larger version. 

Name:	ex 1.png 
Views:	23 
Size:	26.6 KB 
ID:	24065

I own the offer and am direct linking to it (Sign up form - goal is lead-gen)

Offer conversion rate is about 7.76%. I guess if I could increase that to 14% somehow we'd be doing amazing so maybe you're right and I need to change up lander / offer flow.

Click image for larger version. 

Name:	ex 3.png 
Views:	9 
Size:	16.4 KB 
ID:	24067

About $100 in revenue is missing in this photo as it can't be tracked easily through S2S and postbacks.


08-16-2020 09:49 PM #4 stickupkid (Senior Moderator)

Quote Originally Posted by rtking View Post
Would it hurt to duplicate twice? Once without changing anything, and once with changing images only.



Not sure what you would qualify as bottom of cost, but my last 7 day CPA is about $29. Payout is anywhere from $22 - $45 depending on various factors. My ads CPM is $16.83 and CPC (link clicks) is $2.62

Here is ads manager screenshot for past few days

Click image for larger version. 

Name:	ex 1.png 
Views:	23 
Size:	26.6 KB 
ID:	24065

I own the offer and am direct linking to it (Sign up form - goal is lead-gen)

Offer conversion rate is about 7.76%. I guess if I could increase that to 14% somehow we'd be doing amazing so maybe you're right and I need to change up lander / offer flow.

Click image for larger version. 

Name:	ex 3.png 
Views:	9 
Size:	16.4 KB 
ID:	24067

About $100 in revenue is missing in this photo as it can't be tracked easily through S2S and postbacks.
Normally a lead gen offer must be higher than 10% imho. Ofcourse some geo's, some niches its different. Also adding a prelander (since you are direct-linking) to it would lose some clicks, but often raises CR.

What are your ad ctr's in FB?


08-16-2020 10:20 PM #5 rtking (Member)

Quote Originally Posted by stickupkid View Post
Normally a lead gen offer must be higher than 10% imho. Ofcourse some geo's, some niches its different. Also adding a prelander (since you are direct-linking) to it would lose some clicks, but often raises CR.

What are your ad ctr's in FB?
Yeah definitely worth a test with the pre-lander. Been lazy on that one.

The AD CTR is 0.64%. But this is also to cold audiences. Just using a 5% LAA based on past conversions. For some reason the 5% converted better than the 1% in initial tests.


08-17-2020 03:47 AM #6 fbqueen (Senior Member)

@rtking looking at your stats there's a lot of room to improve.

1. Your linkCPC it extremely high and your linkCTR is super low which means you definitely need to work on finding better creatives. You need to have one goal in mind while running on Facebook: getting as much engagement as you can - the more engagement you have on your ad, the better Facebook will like you as an advertiser. You'll get better quality and cheaper traffic when you provide positive user experience. Also when you have lots of engagement, that creates trust in people, therefore the offer will convert better.

Duplicate the adset (which is making you money) into a new ABO campaign and test new creatives in them. Try video ads - Lumen5.com style videos are the best. Depends on the offer, sometimes image ads do really well too. Again, focus on accumulating engagement on them because the more social proof you have on an ad, the more money you'll make with it.

2. Testing not only new creatives but audiences as well. Here's how to go about it: Instead of CBO campaign, I personally like to start with ABO setup, find at least 4 - 8 audiences separated out in different adsets. Each adset should be minimum 1 million audience. This can be keywords, interest targeting, very very broad and lookalikes - the higher the better - the algo will find the right people when you use the pixel and it reports back properly. Make sure the pixel is firing. You gotta aim for minimum 10 conversions / adset / day to keep getting quality traffic.

3. Definitely use an advertorial and make sure load-time is below 2 seconds. It should be congruent with your ad and try to aim for a bounce rate of 55% or less - install Google Analytics on your lander to track this.

Of course there's a lot more to Facebook ads but don't want to overwhelm you, as of now just do these 3 things and you'll see significant improvement.

Good luck and let me know if you have any other questions!


Home > Paid Traffic Sources > Facebook & Instagram