| 9:09 pm on Sep 9, 2002 (gmt 0)|
No. But you can list out the keywords in the order of importance. Just copy all the keywords into your keywords entry box and hit "save keywords" and then "Calculate Estimates".
Check the column under "clicks/day". This should give you an estimate about the number of searches/day and the importance of your keywords.
Else, use Overture keyword suggestion tool. Generally the number of keyword searches will be the same for both Google and Overture (& its partner sites)
| 9:59 pm on Sep 9, 2002 (gmt 0)|
>.Generally the number of keyword searches will be the same for both Google and Overture (& its partner sites)
What math did you use to reach that conclusion?
| 12:17 am on Sep 10, 2002 (gmt 0)|
Sorry. It should read "position of keywords" and not "number of keywords". That was on an assumption that Google users are no different than overture/partner site users. No maths involved here, thank you!
Good way to test the theory is to select some 3 to 5 keywords and enter them into the Overture keyword suggestion tool.
Blue Widgets - X searches/month - 1st position
Red Widgets - Y searches/month - 2nd position
Black Widgets - Z searches/month - 3rd position
Now run the same keywords through Google AdWords. The position of the keywords should more or less remain the same. (of course, the traffic is estimated in clicks/day)
| 2:04 pm on Sep 10, 2002 (gmt 0)|
<<Just copy all the keywords into your keywords entry box and hit "save keywords" and then "Calculate Estimates".
Check the column under "clicks/day". This should give you an estimate about the number of searches/day and the importance of your keywords.>>
What do you mean by that? Where's that "entry box"?
| 2:10 pm on Sep 10, 2002 (gmt 0)|
Create New Campaign
Select language & country
Save & continue
Create ad & continue
If you don't like what ya see in terms of CPC you can then abort :)
I do it all the time when the KWs cost TOO MUCH
| 3:30 pm on Sep 10, 2002 (gmt 0)|
Sorry to disappoint you guys, but there is no way keyword frequencies given by any of the main sources (Google, Overture and Wordtracker) will correspond.
I tried to find out what the 'real' keyword frequency would be for about 1.000f keywords a while ago. I thought of making an weighted average of those 3 main sources. Impossible, because they all have vastly different criteria: Overture aggregates single and plural and misspellings but looks for an exact match. Google doesn't aggregate nor correct for misspellings, but gives 'CTR estimates' based on secret average CTR statistics per keyword. No way to discover the frequency. And Wordtracker results are compiled from the database of dogpile and metracrawler. Wordtracker has many variables and modes (exact, compressed, simple, and so on...) influencing enormously the results. IMHO Wordtracker is OK to discover variations around a theme, and overweights results from the UK, although I don't know how this can be. They are a UK company.
I tested many different algorithms, based on 1.000 very different keywords, from different industries. Results were so randomly changing between those 3 I gave up trying to find out more precise frequencies or rankings.
All three are OK for very rough estimations of keywords and phrases most people will use. In all three, 'computers' will be more popular than 'computer security consultant'. But the ratio between both can vary from 10 to several hundreds!
| 9:45 pm on Nov 21, 2002 (gmt 0)|
I would like to know your results. I'm specializing in foreign language terms, and have analyzed the three sources as well. For a while it was possible to calculate real frequences in Google, (with the previous advertising system). I have noticed that Overture's results increase with time.
One interesting result came from my investigations. Sometimes, some results are distorted by automated searches, (from people who want to know their position for example). On other occasions some results seemed to have been exaggerated by one source. Identifying statistical outliers and constistency between sources was a good trick for me to know which were more realistic estimates of frequency order.