Ads in adgroup not evenly displayed

[MisterXXX]

I always run 3 ads in a adgroup, let them run for a couple of months and then pick the best one. Isn't google supposed to display them evenly, in other words does google give every ad the same exposure?
Because I just checked a particular adgroup which gives me this results:

from Oct 30rd 2004 - Jan 23rd 2005

adgroup 5:
ad 1 39392 clicks 4.5% CTR Displayed 85.5%
ad 2 1008 clicks 4.0% CTR Displayed 2.5%
ad 3 961 clicks 3.7% CTR Displayed 2.5%

another example

adgroup 6:
ad 1 31741 clicks 6.4% CTR Displayed 88.8%
ad 2 1755 clicks 5.5% CTR Displayed 5.7%
ad 3 1240 clicks 5.4% CTR Displayed 5.4%

Or does the ad with the highest CTR always get the most exposure?

[Robsp]

This depends on a setting at campaign level called "Automatically optimize ad serving for my ads". This setting is default on (to maximize clicks and hence G revenue) and Adwords optimizes on the best CTR ad. If you want even rotation, switch this option off.

If you allready have the option off and it is still not working, switch it back on and after an hour or so back off. I've noticed that the optimizer can be active even when the checkbox is off.

Does that help?

[MisterXXX]

Thnx Robsp, the optimizer was switched on.

Unfortunately both adgroups are in the same campaign, one adgroup increased its sales (so the optimizer worked well) and results of the other one decreased (I think that one ad with a lower CTR actually generated more sales than the one with a higher CTR).

I guess I have to add new ads for this adgroup which is performing poor right now and delete the old ads.

Or is there an other way to get around this?

[eWhisper]

There is no way to have partial campaign settings in an AdGroup.

Your two choices are to either move the one AdGroup to a campaign where the settings are off, or look through the following to see how to optimize the settings first.

First, I'd run some numbers to make sure that even though the one ad's conversion % is lower, it's total bottom line because it brings more visitors to your site it's possible the bottom line is better.

Next, I'd make a variation of the high CTR ad to see if I can increase conversion % by using some prequalifying term in it.

And, I'd make a variation of the higher conversion ad to see if I can increase it's CTR while keeping it's conversion %.

I wouldn't delete either ad right away, but add the variations, turn off optimizing, and then split test them all to determine why these ads are causing a difference in CTRs & Conversion %.

For the other group which the optimizer is working well, delete the lower preforming ads, and then make variations of the top one as well and let them both run.

Early in an account's history, I rarely use the optimizer. I want to make a variety of ads, have them all shown, and then determine why I'm seeing the results I am on an even basis - and then start refining ads.

[edited by: eWhisper at 6:29 pm (utc) on Jan. 23, 2005]

[MisterXXX]

Thnx eWhisper, that's useful information!
I am making the changes right now.