| 2:48 am on May 9, 2010 (gmt 0)|
First of all, make sure you have the quality score column turned on so you can see how many of those keywords have decent quality scores. 1700 more keywords? I sure hope they're not all in the same ad group. It'd be pretty hard to write really relevant ads that would cover all 1700 of those keywords.
Make sure you have conversion tracking turned on, and run site query reports to find out what searches your ads are really running on. If you find any stinkers, add them to your negative keywords.
You don't mention if you are using all three match types. Study up on match type.
Keep your best performers in their original ad groups, and take the non performers out so they don't bring down the good ones. Either pause them, or write new ads, but do it in different ad groups.
Finally, don't forget that the landing page is what really converts, not the keywords. The keywords and ads can bring the horse to water, but it's the landing page that's ultimately going to make him drink.
| 1:41 pm on May 10, 2010 (gmt 0)|
Thanks for your reply
I am already doing some of your suggestions
And assuming I do the others
What if in two months I get basically the same results I have gotten in the last two weeks:
KW # Comments ( for the 2 week period)
KW 1-14 Profitable ($25 /conv or less). From 3 to 24 conversions each
KW 15-20 Profitable and 2 conversions each
KW 21-49 Profitable and 1 conversion each
KW 50-68 Paused Cost over $25 no conversions
KW 69-564 Enabled. Cost under $25 no conversions
KW 57-1754 Enabled. No Cost (No clicks) No conversions
I am concerned about the KW in the two groups that are Enabled
I canít afford to let a large number of KW cost $25 each before I decide to pause them
So if the criteria/procedure of:
"If any KW reaches a cost of $25 and has 0 conversions ... Then pause"
Is not good
What criteria or procedure should I try?
| 2:31 pm on May 10, 2010 (gmt 0)|
I can't give you a step by step. You have to come up with whatever criteria works best for your ROI.
| 4:05 pm on May 10, 2010 (gmt 0)|
RE" "Keep your best performers in their original ad groups, and take the non performers out..."
How do you define a "non performer" ?
| 5:32 pm on May 10, 2010 (gmt 0)|
I don't even know how to begin to answer that. It depends on what your goals are for the campaign. You need to know going in what you want to get out of it, and whatever isn't living up to that expectation is a non performer - right?
| 6:04 pm on May 15, 2010 (gmt 0)|
What criteria have you used?
| 8:05 pm on May 15, 2010 (gmt 0)|
Sorry, I think you are asking for more detailed help than I can give you.
| 1:56 am on May 19, 2010 (gmt 0)|
I am not a statistician or an economist.
This is the heart of your problem IMO.
The problem could be re-framed as "after how many clicks can I be certain that the keyword is a non-performer?"
Now a statistician would ask you "What level of confidence do you desire?"
If you can't answer that last question then everything else is just unsubstantiated conjecture and opinion. People will throw around rules of thumb such as "wait 100 clicks" or similar.
And we haven't even looked at the issue of "brand building" or long-cycle sales/ leads.
The more you hang around here you'll understand that people are reluctant to answer generalized questions.