Google is normally very proud of its tools, and has good reason to be.  With a monopoly in the marketplace and easy-to-use tools, they make it easy for new advertisers to find keywords to spend money on, and hopefully grow everyone’s business.

Google has had a keyword tool of one form or another for years.  Back in July, 2008, Google took a huge step forward and announced that you could see average monthly search volumes with advertiser competition details.  Even just yesterday, Google announced updates to their popular search-based keyword tools to be available in more areas.

So why wouldn’t they announce an update like the one that rolled out Wednesday morning?  Instead of simply stating monthly averages it gives you a rounded total of searches for that keyword!  That’s very similar to the old Overture tool that SEOers loved so dearly.  It does this locally and globally!  How cool is that?!

But wait–“local” in this case means your home country, and states the number from the previous month.  “Global” means global, but instead of a rounded number it is an average of 12 months.  This means that it will be very common for your local monthly number to be higher than your global number due to seasonality of given terms.

Why no announcement?  This change should at least be worth a quick blog post. My speculation is because it’s not ready for primetime.  To explain why, I need to let you in on a little budgetary PPC secret:

My biggest and most popular terms are separated into separate campaigns based on match type.  For example, if I’m bidding on the term “shoes”, my “shoes” keyword would have three campaigns that I control separately: 1 for broad match with tons of negatives, 1 for phrase match with lots of negatives, and one for exact match.  I do this so I can be sure to place my largest daily budget emphasis on my most relevant match type: exact.  By having them all in one campaign the broad or phrase match can eat up a lot more of the budget at a higher CPC than my more relevant exact match. (More on this to come in a later post.)

This method has one other huge advantage: you can get some campaign-level reporting detail that you can’t get on a keyword-level report, namely impression share.  The impression share feature tells you how many of the available queries delivered your ad in a percentage.  So, if my campaign featuring the exact match for my “shoes” keyword has 100,000 impressions and a 45% impression share, then by multiplying 100,000 x (1-0.45), then I know that there were 155,000 queries for the term “shoes”.  (Two side notes: 1. These numbers are made up to make the math easy to follow, and 2. You should rarely expect this number to be larger than 90%, due to variances in quality score, daily budgets, and number of competitors).

So what happens when I go to the keyword tool and plug “shoes” as an exact match into the tool?  2.74 MILLION queries.  Now, let me reiterate that my shoes impression number is fake.  I don’t even bid on the term shoes.  However, when I did this same experiment with real keywords, the numbers were similar–as were the results.

Something isn’t right here!  Google’s AdWords reports tell us one thing, yet Google’s keyword tools tell us another.  I hope that this tool begins being reliable one day, because this information can be wonderfully helpful for every level of search marketing, but in its current state I hope you review the data cautiously and don’t make too many business decisions based on this information.