Hi,
I'm new to split testing different "ad" and "lander" variations, but I've been reading up on some of the posts here and on the internet. And I definitely see how split testing different variations works to find a higher converting page. I go over these case studies and I see the variations are sometimes so small but it increases conversions.
Based on your guys experience, is there any rhyme or reason as to why these small variation changes cause an increase in conversions? Or is it just completely random? Do some variations work better one month and another month the same 2 variations are used but this time the other one performs better?
Thank you!
If you give it enough data the best variation keeps being the best for a long time as long as you send fresh traffic.
It's how the human psychology works. Small variations can make huge changes. I have had a 50% increase in ROI in one campaign just by using a black background.
It's about the different perception it gives.
And the reason why you must split test a lot is because the visitors to the page are all different age, different likes, preferences etc. You gotta split test thus sample the audience of the site to get an idea what is the majority demographic and pshycographic of the page/placement.
That's interesting and crazy that such small changes create such a difference in ROI...time to start testing different variations! Haha...
Whats great about it is if you find how to increase your conversions and after it find another way its multiple each other
Lets say your earning EPC was 1$ you found two ways one to play with the bids, one other to increase CTR ad each of them gave you 150% you will find yourself earning 2.25 EPC and not with 2EPC
And sometimes those changes really make the difference between profitable campaign and not. and some of them cant be copied just by click on your ad
My role of thumb:
In general if you think of something that is enough different from what you doing but will not take you much -> time do it today
If its enough different but will take you medium effort -> then do it tomorrow or next week
If its enough different but will take much time -> start to think how to get the same effect but with less time and effort
Each day you don't test new things or challenge your work -> you loss that day.