Home > Questions and Answers > Traffic Source Questions

[facebook question] More demo testing? or more image testing? (9)

05-30-2011 02:31 AM #1 2022 (Member)
[facebook question] More demo testing? or more image testing?

So, while i was working on my facebook campaign today.

i was kind of confused whether to focus more on demos or images.

for an example.

The campaign i was working on.

I was doing...

40 Images for 18-23 (male) demo.

but it didn't do well..

so I'm asking..

would it have been better if.. i did...

40 Images for 20-30 (male) (broad targetting)
= Then pick out the images with good CTR and THEN split up ages for testing?

Orr...

10 (small number) Images of each age groups? such as - 10 img + 18-21, 10 img + 22-25, 10 imgs + 26-30 and so on.....



ARGHHHHHHHHHHHHHHHHHH

Thanks bros!

05-30-2011 02:35 AM #2 stackman (Administrator)

Hey, i personally would do 40 images for males 20-30, find the top 5 images, then split those 5 images up among diff age groups. Usually images translate similarly across all age groups.

Also if the 40 images don't go well and you don't get the CTR you want, don't break them up into small age groups. That's when you gotta pull your pants up and get another 40 images and try again until you get the CTR you want. THEN you split it up across different age groups/

05-30-2011 02:45 AM #3 The Angry Russian (Moderator)

Because of ad decay here's what I would do. Run demo reports in FB and see which ages and interest got you the highest CTR. Target that demo for testing the new images. Once you get a sick CTR ad launch it again on the broad demo.

Note when going CPC facebook is retarted and a winning ad may flop if it doesn't get clicks within the first few k impressions. The solution, launch 10 of the same ad to give them the best chance.

Back in the day (2 years ago) this tactic worked great I would just launch 1000s of ads and once in a while 1 would take off because it got clicked on in the first hundred impressions giving me cheap ass clicks. I wouldn't launch 1000s ads in todays FB nazi regime.

05-30-2011 02:56 AM #4 2022 (Member)

Thanks Stack!!

that's what i was kinda thinking of doing.

going broad with imgs and then split up.

@Russian
didn't think of that demo report...

so if i had 5 winning images, would i have to launch 50? lol
but yeah, i see how that's gonna work...
since i only launch things CPC...


P.S - oh and btw, when i run demo reports. i don't see CTR tab. did fb delete this tab? i only see..

Date - Campaign - Ad Name - Demographic - Bucket 1 - Bucket 2 - % of Impressions.

OH And.. it's weird.. becuase the age doesn't break into groups..

if i targetted 18-24, it only shows me 18-24. no groups....



thanks broski!

05-30-2011 05:49 AM #5 The Angry Russian (Moderator)

Not sure where you're looking 2022.

I was banned six months ago, but looking at one of my clients accounts and see the following.

Click Reports > Responder Demographic > Select Your Filters

I get bucket 1, bucket 2, impressions, clicks, CTR

Under that I get gender, age, country, state VERY powerful info

NOTE: the age buckets are the standard advertising buckets used my most advertisers. You can't break those down further. But if you were to go broad say 16 to 40 it will break it out 16 - 18, 18 - 24, etc

You can also run responder profiles. These are only valid for 14 days, but in here you'll see interests, likes, and so on.

05-30-2011 06:58 AM #6 403flux (Member)

So how many total ad creatives will you guys usually have in one campaign? I don't like to put more than 50 in one and I'll just start a new campaign to test new creatives, do you guys just continue adding more ads into the same campaign? I find it tough for newer ads to get any impressions since the older high ctr ads are eating them all up. Am I doing it wrong?

05-30-2011 08:19 AM #7 2022 (Member)

thanks BROSKI!!!

05-30-2011 10:18 AM #8 The Angry Russian (Moderator)

There's no doing it wrong. Whatever works and makes you money just keep doing it.

My tactic used to be by tagging my campaign with "Wave" paramater. So when I was first testing a campaign it would be Wave1 with 50 to 100 ads... then after I got some learning start a new campaign Wave 2. By the 4th and 5th wave I'd only launch 5 to 10 ads. Then the Waves after that are basically image refresh due to decay.

Also tip if you ever run into this "Awww damn my campaign fizzled out. Guess its over. Well it was good while it lasted" Here's what you do. Pause, wait a month, then run it again. FB keeps getting soooo many users, and peopleget older fall into your age bracket, simply waiting a month and unpausing can give you a quick 1 to 2 week boost of profits.

05-31-2011 02:26 PM #9 The Angry Russian (Moderator)

Quote Originally Posted by constantin View Post
i've heard on WF and other places that pausing campaigns can cause problems with getting impressions...if this is true then how to do people do dayparting?
Everything you hear, especially from wicked fire I would take with a grain of salt. Even the shit you run here, what one person experiences may not be your experience. So why dont you try it for yourself then see if it affects your campaign.

My experience with pausing is that it takes time after you pause for the campaign to actually pause (which can get dangerous if you're unprofitable). Same when you unpause takes about 20 mins to get up to speed as well.

Other than that as long as your pausing and unpausing around the same time I never noticed a huge effect.

Home > Questions and Answers > Traffic Source Questions