Home > Questions and Answers > Tracking Questions

---

Is there a way to split test LPs from a single FB ad? (8)

---

Wonder if I can get some feedback on split testing LPs on Facebook.

To get to the point the question I have is, is there a way to A/B split test landing pages from a single Facebook ad?

I had signed up with Clicktool which seemed to have a way to do this using a no-redirect script on one of the LPs that would rotate the alternate pages in and out of that page without changing the url. It all seemed great, but the company closed down right when I was getting my first campaign going.

I moved onto RedTrack.io. They seemed to indicate they did this too with a no-redirect script, however after clarifying things with them on a group skype call yesterday it turned out not to be the case. You can have the actual landing page in the FB ad, but to test different LPs you need to have separate identical ads going to the different landers.

This would be less than ideal having learned from mation in "Crack The Algo" that FB uses the different ads to test different traffic pools, resulting in different quality of traffic going to the different landing pages. Also, different ads accumulate different levels of engagement in the comments.

But, if this is how it has to be done then so be it. I just wanted to see if this is how it is for the more seasoned hands, or if there is a way to A/B split test LPs from single ads.

Any help is much appreciated.

---

Hello ! You might want to rethink the way you are thinking the different subcategories in fb ads directly.

Under the campaign group you got the subsequent "adset group" and finally the "ads". As inside an adset you can add multiple ads, it's easy to create one ad with the link A and the exact same ad with the link B, that way you will be 100% whitehat.
Another way is also to use the ab testing feature from facebook itself.


Hope it helps and don't hesitate to ask for further information,

Jovana - lemonads

---

To be honest, I think that's how it works in all mainstream trackers right now - direct tracking scripts don't rotate pages in and out of a single page without changing the url, you need to add all links to Facebook to test different landing pages.

By Clicktool, you mean https://clicktool.com? It doesn't look like they're shut down. And you say they supported it?

This looks like a nice addition to our completely redirectless tracking method that no one else has yet. Maybe in the future

Let's see if others have any suggestions how to go around this.


Karolina

---

Hi Jovana,

When FB tests different traffic pools, does it do so at the adset level, or the ad level? If it is at the adset level the multiple ads with variations within the adset would be effective for split testing. If it's at the ad level, there would be different quality traffic going to each ad creating not really comparable results. I'm keen to know the answer to this.

My understanding of the FB A/B testing function is that it's for testing ads rather than LPs.

Ben

---

Karolina, I think I have come to the conclusion you made here. I think I have also learned that a click tracker isn't needed for FB anyway, given their tracking is already pretty good.

Re split testing landers with FB traffic, possibly LP builders like Unbounce or Instapage could be the solution.

Re Clicktool, yes the website is still up, but try contacting anyone. Their training videos went dark about a month ago. For split-testing they had a no-redirect script that you would install on an otherwise code-free LP. The script would swap all of the page variations in and out of the one page, creating the redirectless landing page. I was excited to use it but when it came to installing the script on the page the url was faulty (https://www.null/12345?soandso), and seeking help with it, there was no response. It's a shame, it looked promising.

Ben

---

Google Optimize is an easy and free way to get started if it’s relatively simple changes to the page (and to avoid fb bans, they should be)

Some trackers like funnelflux natively support no redirect rotations. Others like Voluum and Binom need you write a pretty simple script to rotate your landing page versions 50/50

You can see binoms example here, but should work with any tracker

https://docs.binom.org/lp-pixel.php#p1

We’ve actually removed all tracking from all funnels on fb campaigns, definitely helps to avoid bans.

---

Yep you have to be careful with how you do your split testing. It's easy to trigger the circumventing systems rule if you use the server side php example. https://www.facebook.com/policies/ad...enting_systems

The FB pixel scans the page content and sends it back to FB. So they know the content has changed. Depending on who reviews the page, they may or may not trigger a ban/violation warning.

I have a feeling client side solutions like Google Optimize is safer & better, but that's not 100% confirmed yet.

[SSP] Removing all FB pixels from your funnels is one way to avoid bans, is you really want/need to optimize with the FB pixel then the new Prosper202 Deferred PIxel/Pixel Proxy solution is the best option.

It proxy fires all the FB events from a pixel on your own clean domain, and reduces the chances of getting banned because you had your pixel on a 3rd party offer that got banned because of something another affiliate did.

https://prosper.tracking202.com/blog...-s2s-postbacks

---

It's honestly easier and safer to just duplicate the FB Ad and send it to a different landing page. Then run the two ads going to two different pages and compare the results.

If you change or split test landers on an already approved ad, they will probably reject it.

---

Home > Questions and Answers > Tracking Questions