BANS Today

Why KEI Sucks

Written by mbpage on August 15, 2008

When I started to learn search engine optimization one of the suggestions was that you should identify keywords with high KEI.  Is this something you do?  Well perhaps you should reconsider.  KEI is supposed to be a useful indicator of our potential to rank in the serps for a given keyword or keyword phrase yet it has many inherent disadvantages.

Consider this example from my research today “baby”.  Do you think you could rank no. 1 in google for the keyword “baby”?  Not likely, in fact, never in a million years if you’re a one man band yet guess what?  The KEI is 27.98.  I’ve read on many forums that a KEI of more than 1.0 makes a keyword attractive so by those standards this looks a no brainer.

Let’s delve a little deeper though.  If we look at the number of searches we see that there are 159,000 searches a month for the term baby (nice) and the competition is a paltry 901 million!!!!.

So we can see the need to exercise some common sense here.  The competition is way to fierce regardless of the KEI to target this keyword.  What we want is a high number of searches with as low a competition number as we can find.

Take a term like “cat stroller” KEI 27.59, searches 2000 per month, competition 159,000.  In theory this is the kind of term you should be targeting.  From this it is easy to conclude that we need to take all three variables into account when choosing keywords to target.  But there is another problem.

KEI varies according to the data center, keyword program and search engine you obtain the information from - massively.  I targeted a phrase recently where the competition was shown to range from just 400,000 to 7 million, the KEI varied from 0.25 to 75 and the search volume from 2300 per month to 27000 per month depending on where I pulled the data from.

This is the reason KEI sucks and why a good dose of common sense is also needed when evaluating data.  Just try to get the same results I did for the word “Baby” and you’ll see what I mean - you’ve got no chance.

Now I’d love to offer a silver bullet to solve this mess but it simply doesn’t exist.  There is no current solution to this problem and until something comes along we have to rely on common sense and gut feel.  Hardly scientific, but hey, that’s part of the challenge.

In summary then KEI is a useful indicator but no more than that.  You still need to take account of search volume and competition, take the figures with a pinch of salt and test your results.

Page Rank Update - What The ****

Written by mbpage on July 28, 2008

So I’ve just been through my first Google Page Rank Update.  Where is the science related to this?  I have a dozen or so sites that have accquired PR1 & PR2 status out of the 35+ I have developed over the last 4 months.

Why some have been awarded PR and others haven’t is a complete mystery to me.  One site, another blog I have only made 2 posts on in the last 4 months and has no back links and no traffic (now PR1) whilst this site posted to regularly and having a number of back links and good traffic gets no PR at all.

I’m not attached to PR (which is just as well) because it seems completely random.  In fact on of my sites got a PR2 that is only just a month old and hasn’t really been worked at all yet.

What has your experience of PR been - I’d love to hear from you please leave a comment.

The SEO Swindle

Written by mbpage on July 17, 2008

Sometimes you just need to start with a controversial title. So why “The SEO Swindle”? Well what is the purpose of a link building strategy? To improve your rankings in the serps and build page rank right? The higher your PR the easier it is to exchange links and the higher you rank in the serps.

Well I was looking at a friends site today as I’ve been asked to convert it to a blog. I ran it through several tools including SEO Elite and discovered it had only two backlinks. One from the original developer who has a PR5 site and one from a blogger which is a PR0.

The site in question has no SEO and I mean none. Meta title is the company name, no meta keywords, no meta descriptions - zip, nothing zilch. The only positive is the age of the domain which goes back approx 7 years.

So why do I mention this? Will it has a PR3! A PR3 on a site with just 2 links and no traffic. What this says to me is that PR is awarded simply because of the PR5 link of the web developer. The age of the domain may have something to do with it but I can’t back this up with any evidence.

In conclusion if you want to get a PR value for your site simply get a link from a a high PR site with a small number of links that isn’t a directory or article directory. Am I right or am I right wrong?

US Hosting UK Sites (update)

Written by mbpage on July 11, 2008

In my post US Hosting UK Sites I discussed my realisation that my .com and .net domains targeting the UK were not being indexed in the UK and the plans I had to turn things around. I am pleased to say that these have been successful. The solution was similar to that outlined in my post on .info domains.

If you read that post you will know I use HostGator because of the exceptional value for money they provide compared to the cost of UK hosting. The disadvantage of this is that any domain with the exception of a .co.uk will not easily (if ever) find it’s way into the google.co.uk index. With Googles webmaster tools also failing to sort this out their are only two solutions.

Host country specific sites in the country you are targeting (not as easy as it sounds because in the UK the leading hosting company’s servers are located in Germany) and often expensive e.g. £9.95 per month per domain.

A much better and much cheaper solution is to use HostGator and just buy country specific domains. I bought .co.uk for my UK stores via Godaddy. Moved my existing stores and redirected the old domains as outlined in my previous post and low and behold they appeared between 2-5 days later in the google.co.uk index where they need to be.

Google Search Volumes

Written by mbpage on July 10, 2008

I’m sure you caught the news yesterday that Google was now making search volume information available via it’s free keyword tool. This is fantastic news and with the data seemingly verified as accurate by a number of independent experts one wonders where the future lies for the likes of nichebot and wordtracker. Not that I think they’ll go bust but they will certainly need to deliver more to their subscribers now.

I have spent a couple of hours comparing data with the above and some of the differences are marked. Mainly on the positive side I have to add with many terms having higher volume than wordtracker was estimating. The great news is that using the Google tool has been a really boon already for determining niche store categories. I think it will make all of us more money and when used in conjunction with any tool that can help analyzing competition it comes into its own. Can’t wait for a Keyword Elite update myself.

For me the area of most excitement is being able to search keywords by territory which makes it the best international keyword tool available. Think I might build my first BANS Australian store to celebrate.

Googlebot and .info domains (update)

Written by mbpage on July 9, 2008

In my earlier post Googlebot and .info names for build a niche store I discussed my challenges with getting indexed for one of my stores. After 8 weeks nothing. I’m pleased to say I found the solution. Many of you will be aware of this already but those who may have built many stores and think you’ve wasted your time the way out of the mess is as follows:

Buy a .com or .net domain, move the contents of the .info site to the new folder and then perform a permanent redirect on the .info name to help preserve any links you may have built.

Now this may sound technically complex but it is dead easy if you use HostGator (which is one of the many reasons I recommend them for Build a Niche Store users):

You make a copy the existing domain (.info) folder and rename it to whatever you called your new domain (.com) folder.

You delete the files under the old .info folder (but not the folder) and then go to cpanel add-on domains and redirect the .info to the .com. Don’t forget to back everything up first in case something goes wrong.

As a result of making this change my store was fully indexed in 5 days. Yes it’s going to cost $6.95 for a new .com but you should easily recover that in your first month if you’ve targeted a half decent niche.

Related Posts:

Googlebot & .info domains for Build a Niche Store

Written by mbpage on May 31, 2008

This is more of an observation at this stage but I decided to build a store using a .info domain.  There has been a lot of speculation about the value of .info domains and the treatment of them by the mighty google and this is interesting.  The store has been live since 15 May but has still not been indexed by google.  I have built a further three stores since on either .com or .net domains and all of these have been indexed!

Make of that what you will.  One domain isn’t enough to draw conclusions but it adds further to the evidence that .info domains do indeed get treated differently.  Google robot is visiting the site but it currently appears sandboxed.