Welcome to WebmasterWorld Guest from 34.229.194.198

This page requires javascript to function properly. It seems that your browser does not have Javascript enabled. Please enable Javascript and press the Reload/Refresh button on your browser.

Forum Moderators: buckworks & eWhisper & skibum

Message Too Old, No Replies

How to test the effectivness of an ad's copy

comparing copy of one ad to another

New User

joined:May 12, 2005
posts:7
votes: 0

Does anyone have any recommendations on how to test the wording in one ad compared to another.

I would like to use the same keywords as "control" and determine if one ad is yielding a better CTR than another.

Is it even possible to do test this simultaneously, or is the only way to test one for a period of time and the other for another period?

I understand that when you create more than one ad for a AdGroup, AdWords selects the ads to display randomly or alternately. Can anyone verify this?

Thanks in advance

Full Member

joined:Dec 15, 2003
posts:282
votes: 0

You can create more than one ad per adgroup. This is a good way to test.

On the campaign page under edit campaign settings, there is a check box (number 4) marked "Automatically optimize ad serving for my ads." By default this is left on, meaning that your ads will face a darwinian struggle against one another, with the most effective ads (the ones with the highest CTR) being shown more often.

While this is useful for most people, you'll probably want to turn it off for the purposes of testing. This way, the ads will attempt to show themselves as evenly as possible.

While you should get even rotation most of the time, some factors prevent ads from showing evenly when the checkbox is off. Probably the most common is that not all ads have been approved for display on Google's extended network (search sites like Ask Jeeves and AOL, plus content sites like about.com).

I would suggest that, in your test period, you uncheck the search network and content network boxes. This lets the ads run only on Google's site, which means one less variable to worry about. After the trial has run to your satisfaction, you can turn the search network (and possibly the content network) back on, and (if you want) you can turn optimization back on also.

Hope this helps. :)