KEI Calcualtion & Facts

Thsi is a 2 part question.

Firstly is this the correct way to work out KEI???

You divide the Approx Avg Search Volume (from google adwords tool) by the number of results google returns.

Using football boots as an example

1,000,000 / 1,680,000 = 0.595

Is that the correct way to work it out or should you use the allintitle command to limit your results???

Secondly what is a good figure??? I know its the higher the better but what would be a good figure to work with???

Cheers

KEI = search volume/(number of results returned from an exact match search)

although some would say it is searches/number of results

as long as you use the same formula every time you will get the right comparison the higher the number the better the chances you have of getting to the top!

As for a good number to work with well look at the KEI of all your target keywords and take it from there. If there are 1,000,000 searches a month and 1,000,000 search results you have a KEI of 1, it doesn't mean you have a good chance of getting to the top, however if you have 1,000 searches and 1,000 search results you probably have a fairly good chance of getting to the top . I would only use it as an indicator against other keywords but need to look at the bigger picture too. I would also add that I would never use KEI when looking at paid search keywords, I may take it into consideration as a small part of my SEO strategy.

I would ignore it altogether... here's why:

Assumes Wordtracker is using the KEI - to start that number is totally meaningless... don't bother referencing any tools that base their data on impressions especially wordtracker that uses a sampling rate of 0.1% "dogpile search" and then compares that data to a number based on the average market share of the major search engines.

e.g. dogpile gets search impressions [not clicks] for 10 users searching for software that assumes [10 clicks to any one listing whether they clicked or not]

Course no one wants to rank in dogpile so the those 10 impressions are times by 6400 [to get the different between their average market share and Google's... and you will get 64000 visitors uses from that ranked phrase [assume #1 and all search clicked]...

BUT WAIT in which how many are viable visitors?... if every software webmaster runs searchers a few times a day to see where they are and all their SEOs run daily reports how many of those were in the original "10"? 5 or 6? ...let face it... how many "real searchers in your market search dogpile?" ... thereafter the data is really bogus.

add in new competitors looking for content ideas, bloggers, and other information hounds that are not "viable customers" how many of those are in that original "10"? 1 or 2?

In retrospect a phrase that wordtracker says is worth 64000/day is probably more like 6400 and if your research isn't showing 10's of thousands of uses then you phrases may not be a viable term for customers at all.

That doesn't mean "their wrong" - just their guesses are really skewed.

When I was a practicing SEO I never had a client that was satisfied with the holly grail term and being at #1.

There is an enormous difference between paper SEO and practical SEO.

The best tool is hittail or simply any analystic tracking software that shows you data in real or near-real time... statcounter... obviously these need clicks to help you ... but you website is already about "something" - so ignore the holly grail terms and target what you can nail down "today" [for clicks in a few days] rather that months & months of sweat equity only to be disillusioned with the final counterdown.

I really agree with Fathom, KEI data is meaningless.. Focus on getting higher converting keywords instead (looking at your server/site/analytic data log then checking those keywords where visitors mostly buy).