Ad Copy A/B Split Testing Question

Status
Not open for further replies.
M

makethatmoney

New member
Mar 1, 2008
320
2
0
Alright - I have a campaign running and have been working on building it up to hopefully make me some cash. I've been running anywhere from 2-6 ads per ad group so I can see more quickly which ad copy does better generally speaking (positive, negative, question, etc).

Now, I check performance MULTIPLE times/day, but how often should I delete/re-optimize ad copy? Potentially I could do it every click (that would be stupid) but should I let the set of ads run for a week and see?

It would make the most sense to have them run for a week (overall performance instead of daily performance) but I'm not entirely sure.

Thanks all :bowdown:
 


Depends on the volume of the campaign and how good or bad the ads are doing really

You should be able to go a week or two(or less if they bomb) and have enough data to choose a few ads and optimize further from there

Really though you should be able to judge by looking at the campaign. There's no real set rules.
 
Thanks Will - That's what I have been doing. However, sometimes I'll notice after a day I have 10% CTR on ad1 and 3% on ad2 - Naturally I'll revamp ad2, but maybe that is too soon?
 
The only thing I would add is that I never test more than 4 per ad group on the initial test. I typically will write out 6-10 but then choose no more than four. At some point you are just reaching with your ad copy and the quality is going to fall off. I try to create multiple creative concepts and test them against each other. After that A/B test produces a champion then I start testing within the champion concept. It could be as simple as changing one word. Once you test out the concepts you can get really scientific with the next set of a/b tests.
 
Status
Not open for further replies.
Top Bottom