KEI Formula Misleads for Keyword Competitiveness Research

Many SEO specialists wonder why I don’t use Keyword Effectiveness Index, or KEI, to research the right keyword phrases to target.  On the surface, the KEI formula makes sense, and it struck me as so obvious when I first learned about it. 

To the best of my recollection, WordTracker invented KEI, and their original description of the formula was, “the KEI compares the Count result (number of times a keyword has appeared in our data) with the number of competing web pages to pinpoint exactly which keywords are most effective for your campaign.”  What better way to research keyword competitiveness?

At first a fan, I did eventually come to my senses.  This formula tracks how many websites are in a given database for a searched term.  But it is not the volume that counts; it’s the distribution.  Here’s an analogy… 

Which way would you prefer to cross a city on foot:

1. A small alleyway, with a thousand thugs lounging in cafes around the city.
2. An equally small alleyway, with a dozen bloodthirsty thugs in the alley bent on stopping you.

KEI would lead you down the equally small alley…the one with very few keyword phrase competitors, but all right in your way, fighting hard for their high search engine rankings.  Is that what you want?  Of course not.  Keyword popularity is not the selection criteria that matters.  The SEO game is not a democracy…at least not yet, but that’s another story.
I had a sort-of related question from a client today:

Say for instance the word “tennis” was hyperlinked all over the web on all different pages and sites yet the links could be linking to 100’s of different places. Doesn’t that make the word “tennis” more competitive because other sites are trying to use it to increase their chances in trying to get it to show up in the search engines?

On the surface, her proposal made eminent sense, but it’s not the total volume that counts, rather the distribution.  Here was my response to her:

That depends.  If There are a million links with the word “tennis” in them, pointing somewhat evenly to 100,000 sites, the most any one site might have pointing would be, just for example, 20 or 25 links with the word “tennis”. On the other hand, there might be only 500,000 links with the word “badminton” in them, pointing to 100,000 sites, but skewed toward a dozen sites that have been battling it out for top rankings, each with 2000 – 10,000 inbound links with the word “tennis”.  It’s not the volume that counts, but the distribution.

Look very carefully at the top 10 ranking websites for a given search term at your favorite search engine…and how well-optimized those sites are for the keyword, how many inbound links they have, what the quality of those links appears to be, etc.  Don’t rely on the KEI formula or any other web-wide aggregate figures for keyword selection.


  1. Really Nice Article about KEI , because keywords is life of Seo campaign