Home > Paid Traffic Sources > Adult Traffic (NSFW)

Adult Banner Creatives Design (8)

12-15-2017 12:39 PM #1 craig88 (Member)
Adult Banner Creatives Design

Hi, what would be the best way of coming up with banners?

I've read that people advise ripping running banners 100% off sites as they rarely work, which is originally what I was doing.

There seems to be a number of different type of banners ranging from the normal image on left, copy on right, to whatsapp messages to other types of messages too. So far I've been ripping copy from existing banners and coming up with my own images to keep costs and complexities down regarding translations. But even with the standard banners(image on left and copy on right) there is so much variation from copy, to fonts to colours to logo's(snapchat, tinder, whatsapp etc).

Is there a better way of split testing images with various copy/formats/fonts etc?

Also when running a camp what is the ideal number of banners to include? I ran my first camp with about 15 ripped banners and the rep on the TS said I needed to cut it down to less than 4, but not sure if that is best practice.

12-15-2017 01:30 PM #2 matuloo (Legendary Moderator)

There is so many ways to approach this and none of them is the ONLY right one.

One thing is certain though - if you just rip from others, you're not gonna get far ... ALWAYS tweak whatever you rip. It's fine to test ripped stuff too of course, but don't just stick to doing that all the time and only that.

The variations like font, colors, logo ... these usually have a rather low impact on the outcome, so do not focus on it too much. The actual copy/angle and the image ... that's where the magic happens. Put 80% or even more focus on testing these 2 elements.

Less than 4 banners per camp is bullshit ... you can test way more, in case your budget and the available volume allows it. When buying 10 000 impressions per day in some small GEO/spot, you cannot test 15 banners obviously. But take it to a large spot with millions of impressions, and you can easily test dozens of banners in a day.

12-15-2017 08:00 PM #3 craig88 (Member)

Thanks Mat, So what I did was take 10 images and 3 creatives and come up with 30 different banners. Now the problem is I won't ever get near statistical significance unless I spend a LOT on testing. CTR seems like it would be the next best measure to use but I know it can fall short and doesn't mean that much at the end of the day...Payout is near $5 so it's difficult to get many conversions for stat purposes..

12-17-2017 08:07 PM #4 matuloo (Legendary Moderator)

Quote Originally Posted by craig88 View Post
Thanks Mat, So what I did was take 10 images and 3 creatives and come up with 30 different banners. Now the problem is I won't ever get near statistical significance unless I spend a LOT on testing. CTR seems like it would be the next best measure to use but I know it can fall short and doesn't mean that much at the end of the day...Payout is near $5 so it's difficult to get many conversions for stat purposes..
Do not judge banners on CTR, you will lose many good ones this way. Obviously, if a banner has so low CTR that the actual clicks are simply TO EXPENSIVE, you can cut it, but the border is hard to set ... so only cut the very worst based on CTR.

The only way to do it right, is to give it the volume it needs. To save some $, chose "representative" banners from the mix and run just those ... a good way to start would be to take 1 banner for each image for example. And start with the same layout. Then retest the best performers with another of your 3 layouts.

03-04-2018 10:40 PM #5 chocolatepapi (Member)

Quote Originally Posted by matuloo View Post
There is so many ways to approach this and none of them is the ONLY right one.

One thing is certain though - if you just rip from others, you're not gonna get far ... ALWAYS tweak whatever you rip. It's fine to test ripped stuff too of course, but don't just stick to doing that all the time and only that.

The variations like font, colors, logo ... these usually have a rather low impact on the outcome, so do not focus on it too much. The actual copy/angle and the image ... that's where the magic happens. Put 80% or even more focus on testing these 2 elements.

Less than 4 banners per camp is bullshit ... you can test way more, in case your budget and the available volume allows it. When buying 10 000 impressions per day in some small GEO/spot, you cannot test 15 banners obviously. But take it to a large spot with millions of impressions, and you can easily test dozens of banners in a day.
Hey matuloo, you're the man!!! Thanks for the suggestions here as this was a breakthrough for me personally. Would you mind giving me some examples of angles that have been successful for you in the past? Angles that haven't been copied a million times? Only if you are willing of course lol

I would just like to have a starting point for inspiration as I come up with my own angles. I do not want to get too crazy with my angles and burn money though. I kind of need some context in testing angles. Context that is both outside of the box and not similar to what I find using spy tools. I hope that makes sense. Thanks!

03-05-2018 02:30 PM #6 matuloo (Legendary Moderator)

Quote Originally Posted by chocolatepapi View Post
Hey matuloo, you're the man!!! Thanks for the suggestions here as this was a breakthrough for me personally. Would you mind giving me some examples of angles that have been successful for you in the past? Angles that haven't been copied a million times? Only if you are willing of course lol

I would just like to have a starting point for inspiration as I come up with my own angles. I do not want to get too crazy with my angles and burn money though. I kind of need some context in testing angles. Context that is both outside of the box and not similar to what I find using spy tools. I hope that makes sense. Thanks!
The angles that worked well for me were usually connected with common sense reality ...

- for example, I had good results with chubby/bbw girls ... it's hard to believe that slim supermodels need to look for partners on a dating site, but chubby girls with lower egos ... why not, right?

So I came up with this angle that I liked :

Ugly or handsome,
big or small
chubby girls don't care,
they'll fuck you all.

Worked wonders for a while

Just be creative and keep in mind that it has to be believable!

Matej.

03-06-2018 02:37 AM #7 chocolatepapi (Member)

Quote Originally Posted by matuloo View Post
The angles that worked well for me were usually connected with common sense reality ...

- for example, I had good results with chubby/bbw girls ... it's hard to believe that slim supermodels need to look for partners on a dating site, but chubby girls with lower egos ... why not, right?

So I came up with this angle that I liked :

Ugly or handsome,
big or small
chubby girls don't care,
they'll fuck you all.

Worked wonders for a while

Just be creative and keep in mind that it has to be believable!

Matej.
Thanks for this man!!! That helped. Keeping the fact that it has to be believable in mind will be a help for sure as I brainstorm.

03-06-2018 09:28 PM #8 matuloo (Legendary Moderator)

Quote Originally Posted by chocolatepapi View Post
Thanks for this man!!! That helped. Keeping the fact that it has to be believable in mind will be a help for sure as I brainstorm.
Indeed, always think about yourself, how would you react to the ad? Would you see it's BS straight away or would it raise your interest?

I know a good AD the minute I see it, my eye is very trained when it comes to spot ads (after all those AM years), so if something grabs my attention ... it must be a killer ad. Whenever I find an Ad like that, I try to understand why it caught my attention and learn from it.

Home > Paid Traffic Sources > Adult Traffic (NSFW)