I'm new here so I hope I'm not asking a question that has been answered a thousand times. However, I do need a little help figuring out how Google Adwords is working.
I have just redone a client website to (hopefully) enhance conversion rates. Their old website is still active and they wish to do a test to compare the efficiencies of both designs, using google adwords. What I need to do is serve the same ad, but have it go to two different landing pages (one on the old site and one on the new) and then evaluation the differences in conversion rates.
I'm confused as to how to set this up in Google Adwords. I want the two ads to get a equal share in Google traffic. Should I:
1. Set up two separate campaigns, each with their own daily budget (split in two) - how can I be assured that Google won't show both ads in the same search, then?
Or..
2. Set up one campaign with the two ads and turn off the optimization feature - will this give me the 50/50 I need?
Any suggestions would be greatfully appreciated.
Maggie
Option nr 2 is the best one to do split testing. Put 2 exactly the same ads in an adgroup and have one go to the old site and the other to the new one. You indeed have to make sure to switch off the optimization at campaign level to get 50/50 delivery.
Good luck.
One small note, though, that I'm sure Robsp is aware of - but which you might not be if you are a newer advertiser:
Robsp says:
Put 2 exactly the same ads in an adgroup and have one go to the old site and the other to the new one.
To this I would add that the ads should be exactly the same except for the Display URL - which in each ad should accurately reflect the Destination URL of the site it clicks to.
Otherwise, the ad with 'mismatched' Display and Destination URLs would be disapproved when it was reviewed.
Hope that makes sense.
And welcome to WebmasterWorld, by the way!
AWA
Maggie
I have another, basic question for you. My client insists that a new ad campaign takes a period of time for a history to be established before it actually becomes effective. I'm looking on the google ad site and can't find anything relative to this topic. Does anyone know what she is talking about?
My switch testing also blew up in my face yesterday (the first day) because of her hysteria - so I'm not sure if what she is saying has an inkling of validity.
Thanks for any help,
Maggie
If you measure conversions it indeed takes a while before you get conversions and before you know which ad will bring better conversions than others. This depends on the traffic of course (lot of traffic means less time).
Does that make sense?
What I would do is pass a parameter in the actual URL and use a different include file based upon it's value. Of course you'd have to make sure you are able to know where the purchases came from. Not sure if you have access to the right technical resources to make that happen.
