Google AdWords' Search Query Report: The Good, The Bad, and The Ugly

Recently, Google added a new report to the AdWords arsenal: the Search Query Report. In a nutshell, this report gives advertisers a list of all the actual search queries on which their ads appeared, along with useful statistics such as impressions, clicks, conversions, average position, and average CPC. This is data advertisers have been requesting for a long time, but until recently had to pull from other sources, such as log files or other analytics tools.

The Good: It's pretty obvious. This report is a great way to discover new keywords, especially those oft-ballyhooed tail terms that are so hard to find, yet convert like crazy. It's also useful for discovering negative keywords. I just added over 170 negatives to one campaign after poring over this report! (Thank goodness for AdWords Editor!) Best of all, if you're using Google's Conversion Tracker or Analytics, the report includes conversion stats, so it's easy to see which keywords are converting and which aren't.

The Bad: The report makes it plainly obvious that there are serious issues with Google's Broad Match. These issues have been widely publicized in the search marketing forums and blogs, so this isn't news. However, I found some really crazy stuff in my report. All our campaigns are set to US-only and English-only, since we're not authorized to sell overseas. Yet our ads are showing on all kinds of non-English keywords, according to the Search Query report. I even found keywords in Russian and Arabic character sets! Not only are these keywords not relevant, we shouldn't be showing on these searches at all.

The Ugly: First, a little refresher. Most SEMs know about Google's Quality Score, which is supposed to reward relevant ads, keywords, and landing pages. Google makes several suggestions as to how keyword Quality Score can be improved, including the following: "You can also narrow your targeting options (ie, using regional targeting) or matching types (ie, use exact keyword matching). We also suggest adding your keyword to your ad text."

Now, a little story to illustrate just how ugly this is in reality. I have generic ads running for keywords like "magazines" and "magazine" and such that go to our home page. I also have more targeted ads for specific magazines such as "newsweek magazine" with specific ad copy, that go to the page for the magazine in question (the Newsweek Magazine page, in this example).

In the Search Query report, I noticed the keyword "self magazine" (phrase match) in my Search Query Report for my generic ads. Avg position = 1.2, avg CPC = $0.51. OK, that's all fine - we have "magazine" as a broad match keyword in that ad group. Here's the ugly part. I also have an ad group in a different campaign for Self Magazine. The ad has that exact phrase in the copy and that keyword in the ad group as exact match, with a landing page for that magazine (as opposed to our home page for the generic ad).

Minimum CPC for [self magazine] in that ad group? $10.00.

I'm sorry, but I just don't get it! Google tells us that to help our Quality Score, we should use targeted keywords, ads, and landing pages. So I've done that. Yet, in this example, I've been rewarded for following the rules with a $10 minimum CPC. Yet, Google shows our generic ad on "magazine" as a broad match keyword for that same query - and charges us 1/20th of the price.

The ugliest part of all is that the visitor has to then search on our site for the magazine they TYPED INTO GOOGLE in the first place. Making searchers search multiple times is one of my big pet peeves. If they are asking for something specific, we as SEMs should make every effort to give it to them, and the engines should help us out, not hamper our efforts!

Just to back up a bit... All in all, I am thrilled with this report and with the fact that Google is providing us with such great granular data. I'm just a little frustrated by the obvious glitches in the system that are now really, glaringly obvious.