AdWords
1.8K members online now
1.8K members online now
Improve your AdWords performance and boost your ROI, CTR, and Quality Score
Guide Me
star_border
Reply

Split testing: Highest CTR but lowest Conversion Rate

Follower ✭ ☆ ☆
# 1
Follower ✭ ☆ ☆

Trying to figure out how to handle this.  I'm testing 6 different ads in one of my ad groups.  My highest performing ad, CTR wise, is my lowest converting ad.  To make matters worse, my worst CTR ad is my highest converting!

There is nothing in either ad that would mislead anyone.  They aren't a ton different. 

Ad #1 has a 2.48% CTR, but a .58 Conv. Rate.  Ad #6 has a 1.36% CTR, but a 3.08% conv. rate.  Been running this test for a couple of weeks. The total clicks are 188 vs. 130.  Is this too small a sample size to count on this info.

It's so infuriating.  I hate to take either one down. 

Help!

Re: Split testing: Highest CTR but lowest Conversion Rate

Badged Google Partner
# 2
Badged Google Partner
Tommy,

Other things to consider. How many keywords do you have in your Ad group., Are your keywords different match types? How are the ads performing on different devices? Did you use the segment to determine if ads are better by different device as well? Did you check to see if time of day or day of week/ weekend has any correlations?

When you consider the CTR or the Ad you also have to consider the Keyword Term ( or Matched search term that triggered that ad). as well as some of the things mentioned above.

The difference between 2.48 and 1.36 is really not that big a difference, but the Conversion rate was a big difference.

Not being able to see everything in your account and any other influencing data, but I would suggest that from an economical standpoint your AD 6 is clearly the winner in your case.

Sometimes we get to hung up in the X's and O's and for get about the ROI. Conversions should be the be all end all factor. If I convert on a keyword with QS 4 and not on a QS 10, what brings me profit?

The point of your A/B tests are to determine what resonates best with your potential clients. A sample size of 1000 clicks is usually a better sample size if it can be achieved without suffering for the cost..

Also when you test, make sure you know what your test parameters are and what the other influencing factors will be.

Good Luck

Re: Split testing: Highest CTR but lowest Conversion Rate

Participant ✭ ✭ ✭
# 3
Participant ✭ ✭ ✭
Hi Tommy,

That is actually a really nice question as a lot of people struggle with this kind of thing.

There are a number of ways to approach the situation. The first being that you need to realize that the only effect an ad has on conversion is through correct segmentation of audiences. For example:

If I have 2 ads that clearly state you can buy and send flowers on my website, then neither will have a statistical difference in conversion, as the ads are not the conversions themselves.

It is only when your ad is vague and does not direct the correct people to your landing page that it could affect conversions.

So in the case of low conversion rate of one ad, ensure that the people coming to your landing page are the correct audience, if so it may simply be a case of variance.

Another thing to factor in is low volume, it may be the case (and with 100 clicks it very may well be) that you do not have enough data to draw significant results. Now a number of posts usually say 1000 impressions or 100 clicks and this is really wrong as it depends on the data completely, and I have rarely seen significant results with such low volume.

There are a number of A/B testing tools out there, I would suggest finding one you like, or if you are statistically inclined you can work it out for yourself.

But to answer your question, I would run both for awhile longer and if you are confident that neither is segmenting your target market differently then I would take down the low CTR ad when you are confident.

Perhaps run another test once you have decided and use the description if ad A with headline of ad B, running a few tests like this helps to identify what exactly about your ad is resonating to your audience.

It may also help to see what your competition are doing and take notes from them. I would not copy them verbatim as you do not want to become a part of the choice fatigue that is out there.

Let me know of any of this helps or you have another question.

Goodluck

Re: Split testing: Highest CTR but lowest Conversion Rate

Explorer ✭ ✭ ☆
# 4
Explorer ✭ ✭ ☆
HI Tommy D,

maybe this link can help you: http://adcalc.net/split-test/
Your problem is not an exception. I would focus on conversions, but I wouldn't decide with lower than 100 actions happen (clicks, conversions).
Usually I use the the 'rotate even' function. and then decide after 100 actions what to do.