AdWords
6.1K members online now
6.1K members online now
Improve your AdWords performance and boost your ROI, CTR, and Quality Score
Guide Me
star_border
Reply

when do you stop an A/B test?

Follower ✭ ✭ ☆
# 1
Follower ✭ ✭ ☆

I'm running an A/B test with two different ad copies in the same ad group. I get about 3000 impressions per month and a CTR of about 3%.

 

Just wondering if there is a point at which I need to kill the A/B test and move on? In the past, I've used http://getdatadriven.com/ab-significance-test to wait until my test is statistically signigicant. But what I'm asking is if eventually I just give up and say "ok, these ads are pretty much the same CTR". Right now they each have 2000 impressions with 58 and 62 clicks respectively. Or do you wait however long it takes until there is one clear winner?

 

thanks

2 Expert replyverified_user

Re: when do you stop an A/B test?

Top Contributor
# 2
Top Contributor

Hi Birchlore,

 

Thank you for this wonderful question.

 

It is always difficult to decide the data you have is statistically significant to take a decision.

 

Recently found an excellent post by Avinash Kaushik about a solution.

 

http://www.kaushik.net/avinash/excellent-analytics-tip1-statistical-significance/

 

Hope this will give you bit more insight.

 

 

Re: when do you stop an A/B test?

Top Contributor
# 3
Top Contributor

I agree with info;. You can make decisions only based on data that  is statistically significant.

Moshe, AdWords Top Contributor , Twitter | Linkedin | Community Profile | Ad-Globe
Did you find any helpful responses or answers to your query? If yes, please mark it as the ‘Best Answer’

Re: when do you stop an A/B test?

Top Contributor
# 4
Top Contributor

Thank you Moshe for another good resource (link) Smiley Happy