AdWords
1.7K members online now
1.7K members online now
Improve your AdWords performance and boost your ROI, CTR, and Quality Score
Guide Me
star_border
Reply

How to test ad's to determine resulting QS

[ Edited ]
Visitor ✭ ✭ ✭
# 1
Visitor ✭ ✭ ✭

I'm trying to get my head around how to best go about testing new ad's. As QS is such an important factor, it seems it is essential to seperate this for each test. What I'm struggling with is Ad Relevancy.

 

So say I test 2 ads, both have the same landing page to remove this as a factor. We are running this against one exact keyword.

 

We have a QS of 7

 

Ad A has a CTR of 4.5%

Ad B has a CTR of 4.3%

 

Normally you would pick Ad A as the winning ad (not considering conversion etc). What happens if this has a low relevancy determined by google?

 

As relevancy is a factor, how do you factor this in the results?

 

What I'm assuming is that I can pick Ad A but if in fact if the relevancy is less, this can lower the QS and then actually end up lower down the results. This would then end up with possibly less CTR and less traffic.

 

Can we run a test and seperate one ad on the same exact keyword so we can see the direct effect on QS?

3 Expert replyverified_user

Re: How to test ad's to determine resulting QS

Explorer ✭ ☆ ☆
# 2
Explorer ✭ ☆ ☆

Hi Matt,

 

The ad rotation settings takes care of this for you. Depending on which rotation setting you chose AdWords will have the ads compete against each other and eventually only show the most promising one, and presumably the one with the highest CTR.

 

 

Re: How to test ad's to determine resulting QS

Visitor ✭ ✭ ✭
# 3
Visitor ✭ ✭ ✭

I prefer to give my ads equal rotation to have maximum control. Using this method is there anyway to test for relevancy?

 

 

Re: How to test ad's to determine resulting QS

Top Contributor
# 4
Top Contributor

Hi Matt, and welcome to the Community.

 

Are these real figures?  If so I think you're getting too involved in the numbers.  4.3% and 4.5% are pretty much the same thing in AdWords terms and, depending upon the number of impressions these figures are created from, could change from one day to the next.  It would be a fairly weird Ad Group where each Ad had exactly the same CTR - you'd have to have the same Ad Copy, I'd think, and even then I'd bet they wouldn't be the same!

 

Once CTR becomes very close between Ads I'd be looking towards other factors - conversion rates, cost/conversion, performance across devices, times, locations, etc.  Once I'd fiddled with all those and let the dust settle I might find that the CTRs have now drifted apart further so the whole process starts again.

 

Jon

 

AdWords Top Contributor Google+ Profile | Partner Profile | AdWords Audits

Re: How to test ad's to determine resulting QS

Explorer ✭ ☆ ☆
# 5
Explorer ✭ ☆ ☆

Relevance is tested against your keywords. The test is how relevant your keyword is to your ad and the score is what this reflects. The ad itself is judged, for the most part, on CTR.

Re: How to test ad's to determine resulting QS

Visitor ✭ ✭ ✭
# 6
Visitor ✭ ✭ ✭

Thanks Smiley Happy

 

No, these aren't real figures. Just plucked out of thin air. I agree with what you're saying. However,

 

What i'm having a hard time with is if the CTR is the only metric I should be looking at to determine the best ad (regardless of conversion etc).

 

How do I know the Ad with the best CTR, assuming similar conversion rate, is going to be actually the best Ad when quality score is pulled into the equation?

 

Surely we would have a senario where a Ad with high CTR might actually have a low relevancy. Over a period of time the QS would drop because of this and actually perform worse?

 

Maybe i'm overthinking the whole process...

Re: How to test ad's to determine resulting QS

Visitor ✭ ✭ ✭
# 7
Visitor ✭ ✭ ✭
This is what i'm confused on. Revelance is tested on the keyword but we have multiple Ads. How can one keyword have many scores?

I assume relevance is tested on the fly. So a less revelant Ad is show further down results, which then results in a lower CTR.

Re: How to test ad's to determine resulting QS

Top Contributor
# 8
Top Contributor

Matt, have a go at the following experiment:

 

- two ad groups

- one ad in each

- the same one keyword in each ad group

 

Set the experiment at 50-50, so the system splits the traffic evenly. I believe (though I do not have this setup handy) that if you have the same keyword in different ad groups, it will not have the same QS everywhere.

 

As for CTR it is by far the most important factor in QS. As in make or break important. You can even not mention the keyword in the ad text, if you'll have twice the CTR of your nearest competitor you'll shoot to P1 in no time. Because you are the money maker for that keyword.

 

Actually, if you manage to increase your CTR even after reaching a visible QS of 10, you'll see your CPC go further down, all things equal.

Calin Sandici, AdWords Top Contributor | Find me on: Google+ | Twitter | LinkedIn | myBlog
Was my response helpful? If yes, please mark it as the ‘Best Answer.’ Learn how here.

Re: How to test ad's to determine resulting QS

Visitor ✭ ✭ ✭
# 9
Visitor ✭ ✭ ✭

Thanks Adwiser,

 

Yes that is what I was trying to do. Maybe I will take another look.

Re: How to test ad's to determine resulting QS

Explorer ✭ ☆ ☆
# 10
Explorer ✭ ☆ ☆

Calin, wouldn't this amount to the same thing? Eventually, AdWords would favor one ad over the other and use that ad exclusively.