1 Million Dollar Pixel???

lastsamurai

Junior Member
Joined
May 28, 2011
Messages
136
Reaction score
20
Hey guys!

Just wondering...

giphy.gif


If a pixel will only get smarter and smarter depending on the moolah you have available for ads, and,
if you had, let's say, $1M to put into Facebook ads on broad interests and let it rip. You would then eventually have a super-smart pixel, right? This according to years of trial error of the superminds/expert facebook ads marketers you can find everywhere. In fact, where did we get the ideas that a pixel is that little almighty piece of AI? Was it Facebook itself who divulged such info. Anyway, my point is, it is actually a good investment to put as much money as possible on FB ads in the end in order to improve your pixel.
Do you agree? Why? Why not?
Then it would be comparable to doing hundreds of little narrow interest tests. Ideas?
Thanks in advance!:D
 
Asked in reddit as well and got this:

--------------------------------------------------------
There is a huge assumption here that is false.

"will only get smarter and smarter depending on the moolah you have available for...."

That is not true. The pixel only gets smarter if you keep educating it in the right way. This is a key distinction to make because simply running ads under a pixel does not make it smarter, having a successful campaign with successful FB pixel trigger fires is what educates the pixel.

So for example, I could run 1M$ of ads and get 10,000 conversions. Sounds pretty good right? That's $100 CPA. Pretty solid cost, but terrible conversion rate.

If I ran campaign with $10,000 but got 200 conversions. Doesn't sound so good. My CPA was $50. So technically this pixel is more educated and smarter because it is able to get me sales with much less ad spend. (aka the conversion rate is much higher)

Obviously this is just a hypothetical, but it is definitely possible to have a scenario such as the one above if you don't know what you are doing.

To answer your question; like with all platforms it is good to continue investment into your FB campaigns and therefore pixel until you stop getting more of a return. If you have run the FB lead pool dry, pouring more money in won't do anything. But that is the case with every platform.

I hope this helped!
-------------------------------------------------------------

Sounds good to me. Educating it in the right way.
 
From my experience, FB campaigns optimize over a period and then they drop. I assume that the algorithm is not an evergreen optimizer. Like the AI conversational example that FB did between 2 bots, that ended repeating the same over the time because technically it was the best answer... or the other one that the 2 bots ended creating their own optimized language that did not mean anything at all.

So essentially there is a "quality" wave so spending too little is wasting resources, and spending too much also. I think if you can graph every campaign you may clearly see this wave over the time.
The only good thing of FB is that they optimize faster and more adequately than other media selling services. But the range of interests are way narrower. So this are essentially the pros and cons of the service.
 
As AI getting smarter, they have a certain learning curve. Once they reach it they stop learning (if our input constant towards the period of the campaign)
 
I assume that the algorithm is not an evergreen optimizer.

Wouldn't it also mean that the audience is probably running out, or perhaps that competitors are beginning to copy your products? (it happens, probably more than we think?)

So essentially there is a "quality" wave so spending too little is wasting resources, and spending too much also. I think if you can graph every campaign you may clearly see this wave over the time.

This is pretty interesting actually. Do you have some more insights on this?

The only good thing of FB is that they optimize faster and more adequately than other media selling services. But the range of interests are way narrower. So this are essentially the pros and cons of the service.

Couldn't agree more. I have also seen in my tests that Bing can be another interesting PPC option BTW.
 
Back
Top