AdWords
2.4K members online now
2.4K members online now
Use AdWords conversion tracking and reporting to measure your results. Have a question about Google Analytics? Post it here, on the Google Analytics Community!
Guide Me
star_border
Reply

How to split test ads properly?

Visitor ✭ ✭ ✭
# 1
Visitor ✭ ✭ ✭

Hi guys,


I just wanted to know.... 

If I test two ads for an adgroup and I set it up to accelerate ads and rotate ads evenly, how many clicks should I give each ad before I choose a winner? If each ad is given 50% of traffic, how many clicks do I allow for each ad before I pause the losing ad and test a new treatment ad?

 

I am finding that say after 7 days of clicks, each ad gets roughly 50% of traffic, but one ad gets more clicks.
Eg. ad1 = 45 clicks

      ad2=38 clicks.

 

How do I pick a winner? Do I need to test these ads more so that they get more clicks before I choose a winner?

Thanks,

Sean

1 Expert replyverified_user

Re: How to split test ads properly?

Participant ✭ ✭ ✭
# 2
Participant ✭ ✭ ✭
Hi Sean,
Great question and congratulations on setting up your test - sounds like you set up your Campaign Settings correctly. The quick answer is that it varies - I'd say that you want to have many Impressions and Clicks in order to receive statistically significant data. In my mind, 45 to 38 isn't really that statistically significant, so yes, I would test more. Typically when I do evaluations, I look to have several hundred, if not thousands of Clicks in order to make a determination.

One thing you might consider is setting up an AdWords Campaign Experiment (link below). When running an Experiment via Campaign Experiments, AdWords will automatically tell you when your results are statistically significant and it will automatically tell you the winner. Enjoy your test and I hope you get the results you're looking for!

About Campaign Experiments - https://support.google.com/adwords/answer/2385204?hl=en
Monitor your Experiments - https://support.google.com/adwords/answer/2375390?hl=en

Re: How to split test ads properly?

Top Contributor
# 3
Top Contributor
Hi Sean B,

In addition Brian's excellent suggestion, I want to say that clicks are not he only metrics to took at when deciding winner adcopy. CTR, Conversion Rate, conversions are very important. What if you have two adcopies one is having large number of clicks and very less CTR and no conversion. Whereas other adcopy receiving less clicks, great ctr and few conversion. When taking a decision I would say look at all the available metrics. I generally either do adcopy testing byweekly or monthly. If two adcopies are having close metrics, I prefer to run them for few more days.

Hope that helps!

Thanks
Neha
Neha Gupta, AdWordsTop Contributor Follow Me: My Blog | Twitter | Google+ | LinkedIn
Did you find any helpful responses or answers to your query ? If yes, please mark it as the ‘Best Answer’

Re: How to split test ads properly?

Explorer ✭ ✭ ☆
# 4
Explorer ✭ ✭ ☆
Hi Sean,

I always use experiments to test ad copies. As vkore91 and NehaGupta answered your question, I would just ad that I never decide under 100 clicks per ad and use some of tools out there that help me decide ihfthe test was relevant: http://adcalc.net/split-test/

Re: How to split test ads properly?

Visitor ✭ ✭ ✭
# 5
Visitor ✭ ✭ ✭

Hey guys!


Many thanks for your help on this.

Google experiements is something that I will take a look at.

 

Just going back to my example. After say 7 days, my ads have the following data - 

 

Eg. ad1 = 45 clicks

      ad2 = 38 clicks.

 

I know this is low, but I am bidding on brand keywords and the impressions are low. Lets say the two ads, have not converted yet and all I am interested in now is figuring out what ad to keep as its getting the most clicks. If the traffic is spit roughly 50% of the time (I never see ads split 50% each way exactly), & after two weeks, this data is revealed - 

 

Eg. ad1 = 106 clicks 

      ad2 =   85 clicks 

 

What do I do here? Ad 1 has reached its 100 clicks, but ad 2 still has not reached the 100 clicks mark. I am confused as to what to do. Shall I let the ads run a day or two more to see if ad2 has reached its 100 clicks mark before I make a decision as to what to keep and what to pause?

 

If after a couple of days pass, and ad 2 gets 104 clicks and ad1 gets 123 clicks, is it fair to say that ad1 is the  winner? They have both reached the 100 clicks mark for the same period and its clear that ad1 got the better ctr.

 

Do you get me?

 

Thanks guys!