**Split-Test-A-Holic** (19)
*Sry for the giant images, I'm not sure how to re size them*
I intend this follow to help you increase the profits you squeeze out of each campaign. This is what I'm going to do and note it down for you to follow. This particular campaign is profitable and scalable. What I'm going to do is make it stable and long term. To do this we can optimize the pre-sell pages and increase EPC.
Traffic Source: Facebook
This is the EPC to beat: $.406 @ 11k clicks.
There are virtually millions of things to split test to increase EPC. We want to initially increase the EPC as much as humanly possible, it will have a better chance of surviving long term and you make more green in the process.
Here's the start of split test 1.
For the sake of fun I'll give these various split test names. The first test will be called
-Landing Page Background colors. Several colors will be tested.
Here are the new pages in CPVLab Tracking Platform starting out.
--12 Hours in-- (LP 7 & 6 are looking VERY promising) Almost every time I see an LP with Double EPC that LP usually will be the winner. However we do not have enough data to give an accurate analyses yet. I think there will be enough data in the next 12 hours however it could take a few more days for be certain. I will alot 100% more traffic to those 2 LPS that have the highest EPC value.
--18 Hour Update--
--24 Hours in --
Went ahead and dumped the remaining traffic for the day on LP7. Overall CVR is def down today but with LP7 revenue remained higher. Its clear that 24 hours isn't enough. I'm going to just keep it simple and dump 80% to LP7 and 10% to the Original & LP6. tomorrow's data should let us know for sure. After this test is finished I suppose I'll test something about the headline.
If you'd like to see something split tested reach out 
& give it a clever name.
UPDATE: 48 Hours In
This isn't the giant leap in EPC i wanted to see. The peacock is a turkey so to speak. But hell 2 cent increase, I'll take it anyday! haha Ok so just another 24 hours then move on to a better test. FB increased the account spend and should get more traffic rolling.
Looking forward to updates on this one, John 
Are you promoting a dating offer? Can't wait to see the results man!
Are these results from US or some other country ?
Hahaha fucking love the peacock.
Can't wait to see more results. Wow crazy to see just the landing page background color made the difference of almost 3x the conversion rate between your lowest performer and your highest.
Very Excited to see how this thread plays out.
*subscribed*
Good luck!
Awesome stuff bro, Excited to see more updates. BTW on what basis did you decide which colors you would be using? Were all the colors that were used are present in offer page also?
Was just about to ask that. The winning lps...are they similar to the offer in color?
Originally Posted by vipinext
Originally Posted by vipinext
Different colors WILL likely perform really different depending on the country too.
-Doing this same Peacock Split in a different camp for a different country, almost same LP, the Baby blue is doing best for that country, In this test Red seems to be the best.
Originally Posted by pain2k
Originally Posted by Mr Green
lmfao
sounds good, do you do this with all of your campaigns? I believed that making your LP's color same as offer page makes jumping from LP to offer easier for a user and also without him being aware of that. But you proved me wrong... 
Whats next in the pipeline ?
Headlines ?
haha I saw the LPs being tested all over FB for several days. The only way I know what it is, is by the LP colors you posted. I saw them all changing the other day.
Killer Landing page btw. I was curious how that was being allowed on Facebook. Basically broke every Advertising Guideline they have.
Yeah I saw them too, you must have a good cloaker.
hhahaha thats too funny. Yeah that page does alright!! It is pretty risky though. Although they've accepted it for over a month now.
I noticed how, lets say "Simple" your landing pages are, have you tested that against more complex ones? I was surprised the page was working for you but you seem to have the numbers to back it up.
Originally Posted by matthewk
Simple. Simple.
One of my favorite design related quotes:
"You know you've achieved perfection in design, not when you have nothing more to add, but when you have nothing more to take away"
I was trying to put it in a way that didn't give too much info away about his LP...
I just mean things like hero shots and was wondering if the OP had tested them vs his current page.
Originally Posted by rockstar john
This isn't the giant leap in EPC i wanted to see. The peacock is a turkey so to speak. But hell 2 cent increase, I'll take it anyday! haha Ok so just another 24 hours then move on to a better test. FB increased the account spend and should get more traffic rolling.
I'm trying to use this tool, which use an algo for statistical significance, and if I input your numbers for CTR and CR I'll got 'keep testing'
CR (from views to conversions ) : Version 'B' wins with 29.15% confidence.
CR (from clicks to conversions ): Version 'B' wins with 9.25% confidence.
CTR (if we even care): Version 'B' wins with 62.81% confidence.
I'm just trying to understand here, I feel that with a 7k clicks sample conclusion should be pretty accurate.
What tool can I use to make decision for my split test ?
Home > Facebook & Instagram > Follow-along Campaigns