Home > Questions and Answers > General Questions

Split testing budget for banner ads (5)

04-13-2014 11:24 AM #1 deniouz (Member)
Split testing budget for banner ads

Hi guys,

I need some opinion on strategies involving banner ads for POF dating offer. I have a total of 8 images and 8 headers & subheaders ready for testing. My main priority is to test my 4 of my headers & subheaders first.

My budget is around $30 per day which means $7.5 assigned to each banner ad test. Will I get anywhere with this amount of money? Or should I just focus on testing 2 of my headers & subheaders first which would allow me to give a larger portion of my budget to each test ($15 per banner ad test).

04-14-2014 05:01 PM #2 caurmen (Administrator)

Test your images first. Images have the biggest effect on click-through rate, which means that you'll then get more traffic from your header / subhead combos, meaning you can measure their effect on conversion rate more cheaply.

You'll probably need more than 8 images: be ruthless with cutting anything that isn't generating a decent CTR. See part 7 of the Getting Started Guide.

Also, are the header / subhead different angles, or just different wordings on the same angle?

04-15-2014 12:51 AM #3 zeno (Administrator)

Yah main focus on the images.

Are you linking to an LP? The headers/subheaders will make a difference but the LP copy will do a lot more to sell the offer.

Ad copy is certainly important but you need people to take notice of the ad first, both to get clicks and to get ads performing well on the traffic source!

04-15-2014 10:07 AM #4 deniouz (Member)

Quote Originally Posted by caurmen View Post
Test your images first. Images have the biggest effect on click-through rate, which means that you'll then get more traffic from your header / subhead combos, meaning you can measure their effect on conversion rate more cheaply.

You'll probably need more than 8 images: be ruthless with cutting anything that isn't generating a decent CTR. See part 7 of the Getting Started Guide.

Also, are the header / subhead different angles, or just different wordings on the same angle?
Thanks for clearing my doubts. The header/ subheaders are not different angles. They are meant to be exactly the same as the headers & subheaders in my landing pages. Does it mean I would have to make at least 15-20 images with the same headers/ subheaders before testing other variables?

04-15-2014 10:37 AM #5 zeno (Administrator)

There isn't really a golden rule here but I will say yes, test 15-20 images with the same headers/subheaders.

You can vary the headers/subheaders on your lander but try to keep this to only a few, e.g. 2-3 - no point in splitting the data up too much when just starting. You are really fishing to see whether your angle/offer works first, a single headline change is unlikely to make a low performing offer and crappy angle suddenly skyrocket to success.

Once you find decent banners/images you can split test the same images but with different headers/subheaders.

Home > Questions and Answers > General Questions