Ezine Articles
How?

Long Tail Keyword Research

Long tail keyword research is essential. You must put together collections of micro niches, which you can dominate. Each niche needs keywords by which people can find it. You need keyword tools to find those niches out in the long tail of low-competition keywords. You cannot simply compete for the most searched keywords -- by now they have insurmountable competition.

Targeting many niches requires you have a large number of keywords. You need to know the number of searches for them per month in order not to waste your efforts on only a small number of searchers. You need to know the amount of competition for the keywords. Major competition would require major effort, and might make the effort not cost effective. The competition consists at least of all those web pages containing the keyword, but especially those pages optimized for the keyword. You can get all this information for free from Google.

You can use the Google keyword selector tool at https://adwords.google.com/select/KeywordToolExternal as a long tail keyword generator. It suggests a large number of keywords with low search volume but low competition.

You type in a keyword (or several). It will suggest related keywords and give you a spread sheet of those keywords, their search frequencies, and other information. You can download the spreadsheet.

Sort by declining numbers of searches and delete those keywords with too few. They are not worth optimizing pages for. What's too few searches? That is up to you, but I have heard people say they set the limit somewhere between 200 and 300 searches per month (7 to 10 per day).

If you wish, you can reserve those keywords with too few searches to sprinkle into ezine articles. Keywords with low competition may bring the article to page one of a search engine's results.

Next you use the Google search page. It is not usually thought of as a long tail keywords tool, but you use it for two competition searches. Do a Google search for the keyword in quotes to find the number of pages containing those keywords as a phrase, that is, adjacent to each other. The first page of the results gives an estimate of the number of pages containing the phrase. Do not search without quotes -- that counts all pages containing all the words in the keyword phrase even if they are not close on the page.

You can cut the keywords with too many exact matches from your list. Your pages will be lost in the clutter if you try to compete for them. What's too many? Everybody has their own limit, but I have heard recommendations of somewhere between 25,000 and 35,000.

Next, find the number of pages optimized for the keyword. A page is optimized for a keyword if the keyword is the keyword is in the page title, contained in the URL of the page, and the keyword is in the anchor text of links to the page. Again, you can do this step of long tail keyword research by another Google search.

You can tell Google to filter for pages with these optimizations by specifying, for example, inurl:"keyword" to select only pages with the keyword in the URL. (You can find information on advanced Google query operators at http://www.google.com/intl/en/help/operators.html ) Delete the keywords with too much competition of this kind. What is too much competition? Again, it is a matter of taste, but the boundary may be somewhere between 50 and 300. (I'm relying on the opinion of James Jones who suggested these limits.)

Pages simultaneously optimized for the keyword in title, URL, and anchor text indicate that someone is intentionally trying to compete for the keyword.

If you want to sell to the people searching with these keywords, it would help to know if those people intend to buy. You can check their "online commercial intention" (OCI), at least estimates of their OCI, from an algorithm that some people at Microsoft have trained -- there are questions about the methodology and assumptions used in this tool. Give the tool a keyword and it gives you a fraction between zero and one. Zero means the search seems to have no commercial intent whatsoever, and one means the people searching seem intent on buying NOW. The fraction indicates a kind of confidence level in the answer. It does not indicate a fraction of the people intent on buying. In experiments, the average value returned for non-commercial keywords was about 0.2, and the average for commercial keywords was about 0.83. Fractions near 0.5 had a high rate of incorrect classifications. Many people who intend to sell delete those keywords with an OCI less than 0.6 or 0.7. The calculator is on page http://adlab.microsoft.com/Online-Commercial-Intention/

You can do long tail keyword research for free by using the Google keyword selector tool, a Google search, and optionally the MSN online commercial intention page.