Keyword Research Numbers vs Actual Traffic

[kidder]

Does anyone have an idea of what % the different positions in the top 10 result might deliver on average? We have a site that ranks number 6 for what I thought would be a good high traffic search term, I've checked the different DC's to make sure we appear across the board. Google estimates the searches at about 370,000 per month for the US, so about 12,000 per day. Even at position number 6 I would expect at least one hundred per day off this one term but we are not even close to that. On today's stats we are going to get maybe 20 unique visitors. The dominant site in the niche has sitelinks so my guess is it sucks up the bulk of the traffic along with the adwords listings but I am still surprised at the low volume.

[tedster]

If you can get a solid number for postion #1, then the AOL data set that was leaked a couple years back showed these proportions for clickthroughs, over 2 million searches:

#1 - some value
#2 - 3.5x less than #1
#3 - 4.9x less
#4 - 6.9x less
#5 - 8.5x less
#6 - 10.4x less
#7 - 12.3x less
#8 - 14.0x less
#9 - 14.8x less
#10 - 14.1x less

-----

or stated another way:

#2 - 3.5x less than #1
#3 - 1.4x less than #2
#4 - 1.4x less than #3
#5 - 1.2x less than #4
#6 = 1.2x less than #5
#7 = 1.2x less than #6
#8 = 1.1x less than #7
#9 = 1.05x less than #8
#10 = 1.05x more than #9

The trick is getting accurate data for #1

[kidder]

Thanks for that Tedster, your right now the question remains about the "actual" search volumes / CTR for the number 1 spot given the way search looks these days with sitelinks and adsense... The only way I am going to find out is to grab the number one spot and I would say that is impossible. Of course I could work my search volume back the other way and see if that looks about right but it's still guess work.

[tedster]

Also, that data is over a very wide sample. There are pretty strong local fluctuations when you get into specific keywords.

[kidder]

Well I might be wrong but understanding this aspect of SEO is the secret sauce. Does anyone have an opinion on how good the Google external tool is at this point? I still think for any given search Google might have hundreds of variable "top 10" results it can display.

[htdawg]

I also have numerous top 10 positions for certain keyword phrases that according to google adwords keyword tool should give me a decent amout of daily traffic including the phrases that are in the top 3 spots.
Even with other search engines like msn live i have great top spots but hardly any traffic from them.

Makes you wonder if all those keyword tools are any good at all.

[fishfinger]

The numbers appear to be vastly over-inflated from keywords I've checked.

[tedster]

There are many reasons that the tool might be inflated - especially since some queries need to be combined because of spelling, synonyms, stemming variations, seasonal changes, etc, etc.

The external keyword tool is not really there to help with organic search, it's there to help businesses buy ads. But even in ad buying, the numbers often seem to be inflated a good bit. When you think of the often buggy reports in WMT, it's not surprising that this data might have some "castles made of sand" wrapped into it.