Is KEI All It’s Cracked Up to Be

[PeterHoggan]

KEI is an acronym of Keyword Effectiveness Index its goal is to highlight the most effective keywords to target by comparing search frequency or popularity with competing pages. Sumatra Roy is credited with inventing KEI and a slightly tweaked version was later incorporated into Wordtracker. (Keyword services for professional search engine optimization)

By way of example the terms “seo company” and “seo services” were searched 22,200 times and 27,100 respectfully in the UK last month according to Google’s Keyword Tool (https://adwords.google.co.uk/select/KeywordToolExternal). To calculate KEI we divide these numbers by the number of pages returned for an exact match search for each term on google.co.uk.

• “seo company” 22,200/1,530,000 (KEI = 0.0145098039)
  
• “seo services” 27,100/2,580,000 (KEI = 0.010503876)

The higher the KEI, the more popular the keywords are, and the less competition they have, which means they have a better chance of getting to the top. These calculations are based on the Sumatra Roy’s original calculation; Wordtracker tweaked the calculation slightly (P^2/C) or (P/C*P) to put more weight on popularity (P) and less on competing pages (C). Using the new formula we get the following KEIs:

• “seo company” 22,200/1,530,000*22,200 (KEI = 322.117647)
  
• “seo services” 27,100/2,580,000*27,100 (KEI = 284.655039)

Wordtracker tweaked the calculation in recognition of a fundamental failing i.e. the number of competing pages (C) fails to reflect any level of competition. It’s simply the number of pages that contain an exact term match, which in reality tells us nothing about the efficacy of the optimisation used on those pages or what is required to compete with them.

In both sets of calculations above “seo company” emerges with the highest KPI (+38% and +22% respectfully),even though “seo services” is by far the most popular search. Furthermore the calculation still takes into account the number of competing pages, although to a lesser extent. Yet the number of pages “competing” for any term is somewhat irrelevant, what is relevant is the number of pages driving traffic for a term. Regardless of the keyword phrase or how many results are returned for an exact search those pages number 30 and are found on the first three search engine results pages.

So, to calculate the real level of competition you need to evaluate these pages, this is something KEI cannot accomplish and involves reverse engineering and identifying the tactics used by these pages to attain their position. Anything beyond the first 30 results are not competition they’re also-rans.

Nota Bene: The numbers used for popularity in the above examples represent UK local search volumes and were obtained from Google’s Keyword tool. The number of competing pages was derived by conducting an exact match search on google.co.uk (web results). The reason for this was to present figures that are relevant to a UK market.

 

[Employment Law Services]

Great post and very informative........Ive only ever used the wordtracker version and hadn't realised there was a different version out there.

 

[Kevin]

KEI is flawed as it only takes into consideration competing web pages and not links, what the anchor text of these links are, the age of the site and many other factors that are important in determining how competitive a key phrases is.

 

[PeterHoggan]

There is also the danger that if you follow KEI recomendations the terms you target will be pushed down the tail and you miss some of the best searched/converting keyword phrases.

 

[stuarty]

KEI is flawed as it only takes into consideration competing web pages and not links, what the anchor text of these links are, the age of the site and many other factors that are important in determining how competitive a key phrases is.

Eh? It is not flawed at all. It is a formula to calculate the ratio between competing pages and search volumes. Nothing more - nothing less. It is one component of many involved in keyword research. It makes no mention of links, anchor text or the age of a site.

But...I'd love to hear how the age of a site determines how competitive a keyphrase is.

As for the title of the thread - of course it's all it's cracked up to be. All top SEOs use KEI in keyword research.

 

[stuarty]

There is also the danger that if you follow KEI recomendations (1)the terms you target will be pushed down the tail and you (2)miss some of the best searched/converting keyword phrases.

1. How exactly?????????
2. How exactly?????????

Have you actually used KEI for a website before? Have you ever deployed the results from a longtail before? In fact do you actually have any experience of this at all?

 

[PeterHoggan]

KEI does not take into account competing pages only the number of pages that are returned for an exact match search. What are pages that rank in position 100 or 1,000,000 competing for? Certainly not traffic.

OK, the age of a site has nothing to do with how competitive a keyphrase is, but, along with links and anchor text etc, it will greatly affect how difficult/easy ranking for a given keyphrase will be. KEI takes none of this into account.

 

[PeterHoggan]

Stuart, I am too long in the tooth to be bothered with this type of stuff. You would be doing me a favour, and perhaps others, if you were to drop the attitude and refrain from the pointless attempts at one-upmanship. Can’t you see that your belittling and berating responses are not well received? It’s not what you are saying, it’s how you are saying it. I sincerely hope you take this in the spirit intended – friendly advice.

 

[stuarty]

Stuart, I am too long in the tooth to be bothered with this type of stuff. You would be doing me a favour, and perhaps others, if you were to drop the attitude and refrain from the pointless attempts at one-upmanship. Can’t you see that your belittling and berating responses are not well received? It’s not what you are saying, it’s how you are saying it. I sincerely hope you take this in the spirit intended – friendly advice.

Peter - if anyone is trying "one-upmanship" it's yourself. I mentioned KEI in a seperate post then a day later you post a lengthy thread about KEI. You then say something like "terms you target will be pushed further down the tail" which contradicts itself.

The very title itself "is KEI all it's cracked up to be" should have been brought into question in the first place. Yes it is all it's cracked up to be.

I've already said that KEI is one part of a process and you single it out without presenting the bigger picture. It's like saying are the tyres on a car all they're cracked up to be.

All the worlds top SEO companies use this in Keyword research - I'm working with one such company just now on a project and it's being used extensively.

I've used KEI, KOI and KFI (google them) in all the work I've done and it's proven to work every single time! They are then used in further steps in keyword research.

Lastly - it's not about belittleing anyone at all - it's about being accurate. I said in another thread that I only comment or make corrections if I know what I'm talking about - which I in this case I do. I believe that people reading should have the facts so as to make informed decisions.

If you're going to post a thread like this at all then present it in context and produce all the facts. Anyone with working knowledge of it would not single it out but present it with all the other components and this would have been a very useful thread for people to benefit from.

 

[PeterHoggan]

Stuart, this post has nothing to do with anything you have posted previously about KEI, KFI, KOI, or any other kind of fish you would like to mention. Have you ever heard of Narcissistic Personality Disorder. Google that!

You say that you believe that people reading should have the facts so as to make informed decisions, then perhaps you can answer the question I asked previously... What exactly is a page that ranks in position 100, 1000, or 1,000,000 actually competing for? And what part do those pages play in a calculation that aims to output a qualitative score?

I do a fair bit of keyword research myself and use Wordtracker amongst other tools but, in my opinion, KEI is as much use as keyword density. To those top SEOs you talk about, they are welcome to KEI or any metric that uses arbitrary data.

Lastly please don’t insult my intelligence:

"Have you actually used KEI for a website before? Have you ever deployed the results from a longtail before? In fact do you actually have any experience of this at all?"

These are your words, if you don’t think they are belittling and berating you are probably a graduate of the Simon Cowell school of diplomatic etiquette... without the humor!!

 

[stuarty]

Hmmm....I'll leave the Googling to the expert

 

[Employment Law Services]

Stuart, I am too long in the tooth to be bothered with this type of stuff. You would be doing me a favour, and perhaps others, if you were to drop the attitude and refrain from the pointless attempts at one-upmanship. Can’t you see that your belittling and berating responses are not well received? It’s not what you are saying, it’s how you are saying it. I sincerely hope you take this in the spirit intended – friendly advice.

Hmmm....I'll leave the Googling to the expert

Stuarty, I did try and say to you in another thread...........don't let the value of your input be silenced by the negative tone you use in some of your posts.

A long habit of not thinking a thing wrong gives it a superficial appearance of being right. :closedeyes: