AdWords
1.4K members online now
1.4K 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

What happens when keywords are "control and experiment" inside a control-only ad-group?

Visitor ✭ ✭ ✭
# 1
Visitor ✭ ✭ ✭

Hi,

 

We are currently runnin some A/B experiments in our campaigns but we are confused about a certain configuration.

 

The goals of our experiments are to see whether an ad-group performs better after we have decided on some changes to keyword bids and so on.

 

So, we have adgroup A set to control-only, and then we have a duplicate of adgroup A, which we call adgroup B, set to experiment only.

 

Inside both adgroups are a bunch of similar keywords, mostly the same, but with some alterations to certain bids and enabled states in the adgroup B. Very straightforward.

 

My issue is that currently, even though adgroup A is set to control-only and adgroup B is set to experiment-only, all the keywords within either of the ad-group are set to "control and experiment".

 

Is that a problem?

3 Expert replyverified_user

Re: What happens when keywords are "control and experiment" inside a control-only ad-group

Top Contributor
# 2
Top Contributor

Hello;

First , please verify that you correctly set all parameters as instructed in this link:

 

Selecting existing keywords, ads, or ad groups for your experiment

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: What happens when keywords are "control and experiment" inside a control-only ad-group

Visitor ✭ ✭ ✭
# 3
Visitor ✭ ✭ ✭

Hi, yes that is configured correctly.

 

The only thing that I'm concerned about is that the keywords in our control-only adgroups are set to "control and experiment" and the keywords in our experiment-only adgroups are set to "control and experiment".

 

Google doesn't define or make clear what the expected behaviour is in this case, and that's what I seek to understand. So, for example, some people, when AB testing will set the ad-group to "control and experiment", and then set individual keywords inside that ad-group to "control only", "experiment only" or "control and experiment". The behaviour there is very clear. Again, what we have is two almost identical ad-groups, one set to conrol only, one set to experiment only. All the keywords in both ad-groups are set to "control and experiment". What's the expected behaviour?

Re: What happens when keywords are "control and experiment" inside a control-only ad-group

Top Contributor
# 4
Top Contributor

By experiment and control you choose the split between your control group and your experiment 

 

>>"Choose how many of your auctions should be using your experimental changes by selecting one of the control/experiment splits from the drop-down menu. The fastest way to collect results is a 50-50 split, but if your experimental changes are risky or expensive, such as doubling your bid, you might want to have fewer auctions use your experiment. You can't change this setting after you start your experiment."

 

http://support.google.com/adwords/bin/answer.py?hl=en&answer=2375383

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: What happens when keywords are "control and experiment" inside a control-only ad-group

Top Contributor
# 5
Top Contributor

Hello,

 

I would add one more quote from the reference which Moshe has referred here, which I think is important:

 

"Make sure that the settings for your keyword or ad are the same as the settings for the ad group they're in – otherwise, they won't show. For example, if an experiment-only keyword or ad is in a control-only ad group, it will receive no impressions. To receive impressions, the experiment-only keyword or ad will need to be in an ad group set to either experiment-only or experiment and control."

 


Just a thought!

Pankaj

Re: What happens when keywords are "control and experiment" inside a control-only ad-group

Visitor ✭ ✭ ✭
# 6
Visitor ✭ ✭ ✭

Hi, 

 

I'm not sure you understand my question. I have configured the experiment correctly at the ad-group level. Perhaps you misunderstood my query regarding the keywords?

Re: What happens when keywords are "control and experiment" inside a control-only ad-group

Visitor ✭ ✭ ✭
# 7
Visitor ✭ ✭ ✭

Hi,

 

yes, well this is not strictly true. Our keywords are showing, being clicked on etc, and none of them have the same settings as they ad-group they are in Why is this?: because, again, all ad-groups are either control-only or experiment-only, and all keywords are currently set to "control and experiment". 

 

So this is why I ask... Google's documentation is poor in this scenario.

Re: What happens when keywords are "control and experiment"

Visitor ✭ ✭ ✭
# 8
Visitor ✭ ✭ ✭
I agree with you, Google's doc is a bit poor explaining that group in details. But here is a video that should make it clear: http://www.youtube.com/watch?v=Is2iv5A_iAw.
You are using the control and experiment items when you are testing A+B vs B only. A would be exp and control but B would be exp only. Hope it makes sense.
Thanks

Re: What happens when keywords are "control and experiment"

[ Edited ]
Participant ✭ ☆ ☆
# 9
Participant ✭ ☆ ☆

Hello,

 

I do not have experience with Ad group experiments. But try to think logical.

I read this thread and was thinking what is the clue of the problem.

 

And I think the clue is that the Adgroup settings works as a filter on Ad group level.(or not)

 

And as far as I understand from these two video´s it is working as a filter. on Adgroup level.  They imply  this in the first video.  In the second video they do not set all keywords in the experimental Ad groups to Experiment only or Control. They let them as they were at the start (which is control and experiment)

 

And than it is logical if you set an agroup on experiment only, and a keyword in the andgroup is control only, this keyword has no option to run. Which is described in the Adwords Help, as the keyword settings problem.

In the help they write : To receive impressions, the experiment-only keyword or ad will need to be in an ad group set to either experiment-only or experiment and control." (If it is in the control, Then it does not receive impressions)  So in that case the the settings for keyword and adgroup are not the same, but is working.

When you put a keyword in the control. But the adgroup is experiment, than they exclude impressions. Or when the keyword is is experiment, and the adgroup is control. That is what this is telling you.

 

But when you set an adgroup to experiment only, the videos tells you that everything is running to experiment only.

 

AdWords Campaign Experiments: Ad Group Experiments :
http://www.youtube.com/watch?v=wUiHSq0FcGo

AdWords Campaign Experiments - Ad Group Experiment Demo
http://www.youtube.com/watch?v=iOZuFkyZUjw

** I´am learning Adwords and find it very interesting.
** Ik ben Adwords aan het leren en vind het erg boeiend.

Re: What happens when keywords are "control and experiment"

Visitor ✭ ✭ ✭
# 10
Visitor ✭ ✭ ✭
Hi,

I found a link which clearly explains the use of these additional experiment status:

http://www.ppchero.com/ace-your-account-adwords-experiments-and-you/

quoting the part which I hope will solve your query "Green is control and experiment, as in no matter which segment is running this area will be unaffected. This is the option to select for elements that you are not testing or are unrelated to your hypothesis."

Thanks.