At WordStream, we recently released a Free Keyword Tool. Then we told people not to trust it. Sounds a bit contradictory, but let me explain…

One of the most common search marketing mistakes I see people make is in trusting the wrong tools and data points. The best (or maybe worst) example of this is probably keyword research. People are consistently taking specific daily or monthly traffic estimates as gospel. Let's take a look at five reasons this may be the case.

1) Limited Data Sources

The problem with a lot of tools is that even if they pull real search queries, they pull them from limited or narrow sources. If a tool is grabbing a select amount of data from a certain type of source (meta search engines, ISP providers, etc.) you're not getting usage data from your actual target market (searchers on different engines and from different ISPs have different user profiles, and won't necessarily map to the types of searcher you're trying to attract).

For instance Microsoft has some slick keyword tools, but they use Bing data. If you're a technology company seeing 80+ percent of your traffic from Google, you might need to think twice before trusting the Microsoft tools.

2) Estimates Are…Well, Just Estimates

Even if you do get access to a source of data that syncs up to the searches you're targeting pretty well, the numbers you get as "estimated traffic value" are often not that actionable. For instance, in the example above, you could jump over to the Google tool. Problem solved, right?

The image displays keyword suggestions from the Google keyword tool.

In the above image (click to enlarge), we see that a rounded estimate is returned, and the estimate is acutally of all the traffic for this and related keywords. That's a really important point! While it's certainly not a secret nor is it hidden within the interface, many people miss the fact that these are rounded estimates for a portfolio of keywords. You can drill down and get a more specific estimate by changing the drop down above to "phrase" or "exact" match, but even then you're still getting a rounded estimate of all the search traffic Google has for that term. What if you're a small shop trying to figure out how to rank? You can't be sure you'll rank number one for things like "it consulting firm", and even if you rank number one you  won't get all the traffic.

3) Not Enough Ideas

Similarly, what if all of the above ideas are too competitive for your business or website? After all, every single IT consulting firm can get those keywords for free, and there are only around 200 ideas:
This image shows the data you can export from the Google Keyword Tool

We can grab these and another fifty, but the odds of our missing out on a lot of potential opportunities (and being led down a too-competitive path) are quite high.

4) Measuring Using the Wrong Metrics

Another trouble spot is which numbers these tools are giving you. Beyond the basic idea that certain traffic estimates may be inaccurate, tools often give other misleading metrics, such as KEI. KEI stands for keyword effectiveness index. KEI references the number of documents that reference a given term. This is a very imperfect measure of competitiveness, because ultimately you're not competing against every mention of a keyword: you're only trying to outrank the top ten results! Using a tool like SEO for FireFox to look at a lot of different numbers (number of links indexed by Yahoo! number of quality links like .edus, etc.) to measure how competitive a word is is really a much better (if also imperfect) means of discovering what competition may look like.

5) The Data Isn't Yours!

This is the biggest one: ultimately these estimates can lead you down the wrong path because they reference general data. The problem here is that just because someone else can rank for a term doesn't mean you can, and the fact that other people are driving tons of traffic doesn't mean that even if you do, it will convert for you.  Ultimately the best way to see if a keyword is right for your campaigns is to test the data on your own site to see how it performs.

So Why Use (Or Build) a Keyword Tool, Then?

Keyword research tools do have value, despite the above disclaimers. Much like services like Compete.com, they offer suggestions for next steps that are directionally indicative. We built our tool to try to help people just getting started find ideas for their industry or niche. While we tried to address some of the concerns above, the tool should still ultimately be taken with a "grain of salt": the real value of keywords isn't having a nice ordered list. It's acting on data and turning keywords into traffic and conversions.

Tom Demers is the Director of Marketing at WordStream Inc. WordStream offers keyword research and keyword management solutions for SEO and PPC. You can catch up with Tom by sending him an Email at tdemers at WordStream dot com, following him on Twitter, or by stopping by the WordStream Internet Marketing Blog.