Keyword Research: Making Effective Keyword Choices

An often overlooked, but vitally important, factor of search engine optimization is keyword research. All the optimization in the world will do you no good if you are optimizing for the wrong keywords. Very early in the planning process keyword research can even help you decide on a site title or domain name. How would you like it if you built a site around a certain domain only later to find out that had you chosen a different domain you could be getting 10x more traffic? Such a scenario is not impossible, so read on.

Keyword Popularity

The first aspect of keyword research you must undertake is to determine which keywords are more popular. There are two main tools that webmasters use in order to do this, the first is called Wordtracker which is a paid service, although they have a free trial that does enough for most of your needs. The second is Overture's (Yahoo Search Marketing) Keyword Suggestion Tool.

Wordtracker is primarily powered by statistics from Dogpile.com, a meta search engine with a small market share (but still a sample size of millions). Overture's results come from Yahoo, which is of course much larger.

Which service to use, and in fact you could use both, is a subject of personal opinion mostly, however I believe Wordtracker to be better. Wordtracker may have a smaller sample size, however when you're dealing with hundreds of millions of samples you're going to have statistically significant results regardless of markershare. Additionally Overture lumps singular and plurals of the same word together, a process called stemming, which isn't helpful at all. Finally, Overture's results can be skewed by software that repeatedly checks their system for current PPC bid prices.

Diary

I once optimized for a singular form of a word and reached #1 in the search engines for it, I was quite happy. Then I did keyword research (which I should have done first) and discovered the plural forum of the word receives 5x more traffic, and so I could have been making 5x more than I was currently. I made the change and eventually got to #1 on both versions, but I ending up missing out on probably around $10,000 worth of income because I didn't do my keyword research first. Also, because of how Overture stems, had I done my research with them I would not have picked up on the problem. This is why I recommend Wordtracker.

Whichever system you end up using, it is important to note that these services cannot accurately predict traffic levels. They may give estimates for daily or monthly searches, but those estimates are almost useless. Instead, these services are really only worthwhile when doing comparisons. For instance they cannot exactly predict traffic for any one keyword, but they can fairly accurately show proportional popularity among different keywords. So while you may not know how much traffic you will get, you will know which keyword has the best chance of providing the most traffic.