Keyword research is more of an art than science. However, with the abundance of both free and (very) expensive tools, you can be misled into quite the opposite belief, especially by some marketers who want you to believe that their keyword research software will catapult you magically into the first page of SERPs and make you tons of money.
Keyword research is important and it can be one of the key factors in the success of your website/blog, if done properly. However, there are many pitfalls that you should avoid. Here's my personal list of the top 10 mistakes that many aspiring website owners are guilty of (including myself).
1. Not doing keyword research at all
That one is self-explanatory. If you give up on keyword research, you miss one of the cheapest and most efficient market research methods ever created. You will never come up with all the ways that people may be search for content in your niches, you will not discover all the simple ways you may get more visitors and finally, you will simply not think of many content ideas that are inspired by what you find in the keyword research data.
The conclusion is - you can brainstorm keywords on your own, but you should do the comprehensive research for your niche at least once and store the keywords for future reference. Much better would be to use some keyword tools on every-day basis - this is highly recommended if you're serious about your online projects.
2. Taking keyword counts at the face value
It always astonishes me when people get some numbers from any keyword research tool and then expect the exact number of visits to their website for that phrase, if they make it to the top of SERPs of course.
The truths is, all the keyword tools have access to a very microscopic sample of all the queries that people enter into the search engines every day. No keyword tool has access to data from Google, Yahoo or even MSN. It means they are relying on some second or third-tier data sources and any estimates of the keyword popularity are just that - estimates.
The more you dig into any niche, the greater the potential discrepancy between estimated and actual search counts may be.
The best I can suggest is to look for relations between any counts provided by these tools. If for example Wordtracker says that ipod is searched for 28,000 times a day and zune is searched for 8000 times a day, there is a good chance that ipod is a much more popular keyword to target (however, as I will show you in a separate articles, different tools may suggest very varying proportions between different phrases).
3. Relying on a single tool
It's always a great idea to test your keywords using several keyword services.
First of all, you will simply get more phrases if you combine results of for example Wordtracker [aff.] and Nichebot [aff.] and the Google keyword selector tool.
You will also be able to eliminate any odd results, which may be a result of your competitors spamming one tool or the other.
4. Relying on over-inflated data
This applies particularly to any keyword research done with the Overture keyword search tool and any other online or desktop tools that rely on these data (such as the very handy SEO Book Keyword Suggestion Tool or the freeware Good Keywords software).
The Overture tool is extremely abused by marketers doing their research and so it shows strong bias in certain niches and unnaturally high search counts (which change a lot - in one month it can be hundreds of thousands for one keyword and just mere hundreds for the same keyword next month).
You should always cross-reference Overture results with another source.
5. Focusing on just one keyword / phrase
The days when you could write SEO-optimized content by focusing on a single keyword / phrase and just taking care of its proper density are definitely over. We are now in the latent semantic indexing era.
This basically means that whenever you're doing keyword research, you must think about a set of related phrases and words that will allow you to theme your page and "convince" the search engine that it's on a given subject.
There are many great ways to discover words that belong to the same semantic category. You can for example try the free WordNet tool. This is a browsable database of keywords where humans define their meanings and relations between them.
6. Relying on KEI
If you don't know what KEI (Keyword Effectiveness Index) is, you may find this detailed explanation helpful.
Basically, it's a ratio of how many times a given keyword is searched for to the number of competing websites that rank for this keyword. Many people rely on this ratio to judge how competitive a niche is and what their chances are to achieve top rankings in SERPs.
But guess what, KEI can be extremely misleading.
First of all, you're really competing only against the top 10 (or 20) listings in SERPs. It doesn't matter if there are 100,000 or 100,000,000 other results (and that would change KEI drastically). The only thing that matters is how strong SEO-wise the top 10 results are.
On the other hand, if you take into account how wrong the keyword tools may be in estimating the counts (especially in case of of the long-tail keywords), you get another reason not to rely on KEI (because doubling the count may not mean much, but it may alter KEI drastically).
There are many free tools which are much more useful in estimating your real competition for any given phrase.
If you're using Firefox, you may check these great extensions - SEO for Firefox or SeoQuake. They will show you a number of useful data directly in the Google search results (such as PR of each page in results, number of backlinks, age of domain etc.). With these data, you can get a good feeling of how difficult it will be to rank for your term.
You can also use a very cool freeware application called Keyword Explorer which will immediately show a number of search results for a normal query and with modifiers, such as "intitle" or "inurl". Usually, even if there is a large number of results for a regular query, but a much lower number for queries with modifiers, it means that there may be not too many pages optimized for your phrase and you have a good shot at ranking well.
7. Not using tools to organize and manage your keyword lists
If you're serious about your keyword research, you'll quickly end up with lists containing thousands of keywords for almost any niche. You can try and use Excel to manage them, but that's, well, not the optimum way. It's a bit a like reinventing the wheel.
There are specialized tools, which will make managing your lists a breeze. You will be able to do all basic operations, such as sorting and filtering, as well as more advanced, such as automatically discover subniches, find keywords for your LSI optimization (see point 5 above), or conduct multi-level, multi-criteria filtering.
One tool that I truly recommend and use almost daily is the Keyword Results Analyzer [aff.]. I am going to have a review of this tool soon, but in the meantime, I strongly recommend that you go to the link provided and sign-up for the free 8-day video tutorial. Not only will it show you how to use the tool, but you will also get a ton of useful tips on keyword research in general.
I have no doubt that my keyword research has become much more efficient since I started using this too to manage my lists.
8. Researching at the wrong time
This applies in particular to people using the Wordtracker tool, which stores keyword statistics for the last 90 days.
It means that it would be a rather bad idea to research Christmas gifts in July using Wordtracker, as its data from the last 90 days would contain very little Christmas-related terms.
It's less important if you're using tools like Nichebot and its Keyword Discovery database, which generally covers a much longer time-span.
9. Going after the most expensive and competitive terms
Unless you're very experienced in SEO, have good contacts in the industry and deep pockets, there is little sense in going for the most competitive and expensive phrases. Your business model is in the long tail - less competition and less income per click may still translate into greater revenue than the most profitable keywords with mad competition.
However, it's a very sound strategy to use the long-tail terms to drive traffic to your site and then to channel that traffic to other pages in your website, optimized for example for high-paying AdSense ads.
10. Letting your keywords lists gather dust
This is something I am very guilty of - I have an infinite number of keyword lists and niches that I researched just because discovering niches can be so much fun, particularly with tools like this one to automatically discover new and profitable markets. But guess what - I've never used these keywords and probably never will. And they will never make me any money.
So please try to follow this little bit of advice - when you have your keywords, make sure that you actually use them before embarking on another project in a new and unrelated niche.
Do you have any tips on the most efficient keyword generation strategies? If so, please share them in the comments for the benefit of everyone!
This content was originally posted on http://guidetomoney.blogspot.com/ © 2008 If you are not reading this text from the above site, you are reading a splog
If you have something like 15 minutes to spare, you could use that time to find some really excellent domains to register. But please note that domain-hunting can get addictive. You've been warned... :-)
If you've been following this blog, you probably know that I like to develop some custom tools to help me with generating keywords or niche ideas. 
.
