I want to share with you an email I send to a group of people that were asking in a private forum about keyword research and how to do it properly. Please excuse the type of language used, it was in an environment of trust and between peers. I think you will benefit from it and that is why MDZ has allowed me to post it here:
Keyword research is an art, to be honest. But you can use the tools that most people use, such as the google keyword tool, keyword elite, and other tools. That is good for the average person, but there is a lot more to it than that.
Disclaimer: When I seat down with a client, I tell him this, give me any keywords you want to rank for, does not matter competition, does not matter the market. That is because I am confident on what we do so I do not care how competitive the market is or who is in that market. This is not the best approach and this should not be the way most people should go about doing SEO. I need to make this clear. Now…
about finding keywords:
There are 3 ways most people would do this:
1) Regular way (most people do this, including many SEOs):
a) go to the keyword tool and enter a 2 to 3 keywords.
2) Some people including good SEOs would do this:
a) go to the keyword tool and enter one word keyword, make sure it is one, not 2 or more words. Why? you want the tool to give you the ideas, not you telling the keywords were to go.
There is a lot that I do in GA, but just to give you some ideas:
I look for keywords that have a lot of traffic, converting well? yes, great! where are they ranking? lower than top 10? crap!!! I have to work hard on these because if they are bringing enough traffic and converting well then by moving them to top 10, top 3 I will make a killing.
I look for keywords that have little traffic but converting well, once again, are they top 10? if not, I do SEO to rank them top 10 because I know that will increase traffic and since they convert well I will make a killing.
I look for keywords that have a lot of traffic and converting badly, are they ranking well? no, then do SEO, yes? then focus on conversions! meaning, look at how long people stay on a page, may be the call to actions is weak, may be my offer sucks, may be the content is bad, etc etc.
These are some of the things I do, it works, trust me.
3) the 3rd method is the one that very few people use, I only use it for my top clients because it takes more time and money, but gives me the best results. I will not explain it here (sorry for that) because it is a technique I developed in the last 9 years and I share it once and was leaked to a few people. In SEO is good to have an advantage over your competition.
But, the first 2 techniques will help you alot, in most cases that is all you need to do.
Oh, and there is more:
4) PPC: yes, PPC as a keyword tool. Create a campaign with some generic keywords, some phrase keywords (never exact) and let it run for a few days (weeks?, a month?) depending on your budget. Go to your GA tool and see what keywords are coming to your site via these generic keywords. I think you still need to use a trick (set up in GA) that will allow you to see the exact keywords that came to your site via the generic keywords.
What do I mean? let’s say you have a keyword soccer on your PPC group.
In GA you will see keywords such as soccer shoes, where to buy soccer shoes, soccer shirts, etc. There you have them, GA will give you the keyword and the daily traffic for it.