Like many folks, I have been using KEI for some time now to determine what keywords I should target with my web site. And this has led me to becoming concerned with results KEI provides and keywords it suggests. I need to say here that my concern is very subjective as many folks are happily using KEI and don't seem to have a problem with it.

My main concern with KEI is that, by way it works, it strongly favours demand numbers without, I feel, sufficiently taking into account corresponding supply numbers.

I need to say here that I interpret supply numbers as a representation of how competitive a keyword is. For example, if keyword 1 has a supply of 200,000 while keyword 2 has a supply of 5,000,000, then I would consider keyword 2 as being more competitive than keyword 1.

And all things being equal, I would prefer to target a keyword that is less competitive and with less demand, rather than a highly competitive keyword that has a higher demand. The reason for this is that I feel that I have a better chance of cornering a section of a less competitive market than I do that of a highly competitive one.

Based on my concern with KEI, I have decided to create an alternative. I have called this alternative "Competition Indexed Demand" (CID). Now, CID works out marketing potential of keywords in a similar way to KEI but it uses a different formula, one that takes more into account supply numbers of keywords (or their competitiveness).

For example, using "ranking" as starting keyword with Overture, KEI suggests following top 3 keywords,

Keyword Demand Supply KEI nfl quarterback ranking 43,474 75,800 24,934 nfl power ranking 43,171 122,000 15,277 college basketball ranking 71,149 541,000 9,357

while CID suggests following top 3 keywords,

Keyword Demand Supply CID dick vitale college basketball ranking 16,983 33,400 640 nfl quaterback ranking 43,474 75,800 427 vote nba power ranking 3,129 30,200 394

Many industries such as travel, pharmacy, pornography, and gaming have search results which are hyper competitive and require heavy advertising or aggressive SEO techniques. Some niche websites may see an even greater ROI on smart link rentals since many of their competitors may not include link renting in their online marketing budgets.

Some rented links provide great value in direct targeted traffic, whereas some other links provide greater value from effect they have on search relevancy.

Most links are rented on a monthly basis with an option to renew at end of month. Some link prices can be as low as a few dollars a month whereas some can cost thousands per month.

There is no singular one-size-fits-all way to directly assume value of a link. Most effective marketing has risks associated with it, but you can minimize risks and maximize your return by breaking value of link down into its elements:

1. Direct traffic from link renting 2. Viral effect of advertising 3. Effects of link rentals on search relevancy.

1. Direct Traffic from Link Renting:

Link renting can drive significant direct traffic. The first things you need to know are:

* what is quality level of traffic? * how related is this audience to my product? * how much traffic does site receive?

What is Quality Level of Traffic:

Where does bulk of their traffic come from? Someone recently asked about ePilot, pay per click search engine, at Search Engine Watch forums. Andrew Goodman replied "The first question I would ask myself would be: 'where does ePilot's traffic come from?' Answer isn't clear? Then it would be no surprise that so-called traffic doesn't convert to anything."

This same concept holds true with any site. If you do not see any quality inbound links that could indicate a problem: or they could get most of their traffic from pay-per-click ads. Whatever case, if you are going to spend money renting links for traffic, make sure you know where their traffic comes from.

How Related is Audience:

It will be somewhat of a guesstimate as to how exactly how related traffic will be, but generally it is fairly obvious when something is on theme or off theme. Some things you may want to consider when determining traffic quality:

* What is mood / mind set of their traffic? Related traffic may not be traffic that converts. At 2004 Webmaster World Las Vegas Conference a speaker mentioned that a website about prom hairstyles may not be a good place to advertise prom dresses since girls do not usually look at prom hairdos until after they have already purchased their prom dresses. * What product should I advertise? Sometimes it makes sense to send traffic to your home page while other times it is more logical to advertise a specific product because it is hot or more related to that particular audience.

How Much Traffic Does Site Receive:

Most honest webmasters who sell ad space understand that traffic they receive is a large part of their ad value and should have no problems giving you that information.

Some systems such as Alexa also track traffic, but their statistics can be inaccurate due to their limited distribution.. The best way to use such systems is not for traffic estimates but for comparisons between sites. Note that Alexa has a heavy bias toward webmaster resource type sites. Additionally other companies such as Hitwise may provide more accurate traffic statistics.