Home > Paid Traffic Sources > Facebook & Instagram

---

Image/headline testing (10)

---

I keep reading the best way is to place 3-4 duplicates of one ad into one campaign to test that image/headline. But this way takes foreveeeeeeeeeeeeeeeeeeer to test a bunch of images.

I've trawled through the majority of the posts in this Facebook subsection and only found mentioned a few times of people loading up a ton of ads into one campaign and seeing which ads get early clicks to take off. Some ads may be ignored entirely without a fair chance, but it's quicker to see if anything sticks.

Thoughts?

---

So are you saying that it takes forever to test 3-4 ads or what?

I'm doing POF at the moment but my "rule of thumb" is that if there's less than 3 clicks when ad reaches 1200 impressions, pause it. Of course if you want to get fancy, you'll check conversions first and decide based on that.

---

I know PoF, you can load up as many ads into one campaign as you want and run them simultaneously if you desired.

I don't think Facebook is regarded the same way.

---

How can testing 1 image/headline in 1 campaign be the most efficient way to test, so in order to test 100 images, you make 100 separate campaigns with that ad duplicated within each campaign multiple times?

Am I missing something?

---

Hey mscimitar,

I am not a Facebook expert but as I know, you really have to create 100 campaigns to test 100 images. That's because if you use 100 images in 1 campaign, Facebook will only give impressions to the high CTR images after certain number of impressions.

Please correct me if I am wrong.

Hey qhead,

You paused after 1200 impressions and require 3 clicks!? That's quite harsh haha. That would mean 0.25% CTR! Anybody did it the same way?

---

Is this not a good thing?

---

If you throw 100 images in one campaign they will get reasonably even impressions until an advert or two get clicks, then they will dominate all impressions from then until the end of time. Regardless of CTR. If you don't reasonably split the images into separate campaigns you might as well not test multiple images at all. Hell, don't even bother splitting by age group, gender or country! One has to learn how to play the traffic source game. For FB that equals tedious, granular campaign creation to mitigate the platforms delivery system. It sucks, but that is the state of play.

---

Yeah exactly. At least that's what I encountered when I tried FB. One ad got a few clicks then all impressions are given to that ad. All the other ads cannot receive any impressions. Then that ad dies quickly....

---

That's not a good thing because FB is optimizing way too early. For example, in a campaign with 10 ads, FB will stop giving impressions to other ads when 1 single ad gets 1 click. However, the other 9 ads may have only received 10-100 impressions! In other words, you are not giving enough impressions to other ads to test. At least we need 1000 impressions per ad to decide to keep or cut it.

---

I put up 20-100 images with the same headline and copy.

Certain ones will take off and then I'll lower their bids to a place where they stop getting impression and let the other get traction to see who's the true winner.

Then I'll cherry pick a few of the best and start testing headlines which is typically going to have some influence on ctr as well.

Then lastly, your body copy pretty much drives the sale. If you're ctr is good but your conversions are bad focus on the body copy.

Hope this helps.

---

Home > Paid Traffic Sources > Facebook & Instagram