Aaron Wall wrote a good blog post a couple days ago titled Why Do Keyword Tool Search Estimates Vary So Much? It's a question that Rand and I have grumbled about many times. Actually, I'm pretty convinced that it's like a rite of passage--you're not a tried and true SEO professional until you've played with at least four different keyword tools ;)

Aaron reminds everyone that the tools should serve as a way of identifying trends in various keywords, rather than be relied upon for hard numbers, and he's absolutely right. You really don't need to know the exact number of searches for any given term, right down to the single digit. When it comes to doing keyword research, a pretty good idea of the number of searches for any given query should suffice.

I think the potential harm here is relying on one tool for a set of numbers. No one tool is proven to be significantly more accurate than the next. When I do keyword research, I usually use Overture, Wordtracker, and Keyword Discovery, and then in my report I include the average of all three tools' predicted queries in addition to their individual numbers.

Obviously, knowing the exact number of searches for a term can result in good dinner party conversation. I can see it now:
Rand: Mystery Guest, did you know that last month 268 people searched for the term "Rand Fishkin?"
Mystery Guest: That's really interesting. Can you tell me how many people searched for the phrase "Shut up and pass the pepper"?
Rand: Aw, shucks. Yes dear...
Hmmm, on the other hand, maybe knowing the exact number of searches for a term isn't as interesting (or as healthy) as we thought.

Seriously though, I wonder how long it will be (if ever) before a keyword tool comes out that is so accurate that SEOs' heads will explode with joy.