This is the ultimate showdown. Well, not really, but let me suspend reality for a minute or 20 as I put these ‘official’ keyword research tools up to the test. The purpose of this post is to try to gain a better understanding of which keyword tools to use and what to expect from them. After all, we all want to be able to effectively pick the ‘right’ keywords from the start and we all don’t have Goldman Sachs size wallets to run a blank check PPC campaigns.
The Setup:
Over the years we have helped many clients achieve and maintain top rankings for a variety of important keywords. For this evaluation, I choose some keywords I know have been in the #1 spot on Google for the last year. Using this as the control and drawing the number of visits for each of these keywords from Google Analytics we are able to see just how close we get to the proclaimed ~42% click through rate. The proclaimed percentage for being in the number 1 position. I gathered data for 14 different keywords. I know what you’re saying to yourself, that is a small sample, and yes you are correct. However, I say to you though that it’s a start. I will not divulge the actual keywords themselves in this post, but I will tell you a little bit more about them. Out of the 14 keywords, they come from 4 different websites. Additionally, 3 of the keywords are brand related. I included this into the mix as to see the difference in click throughs and somewhat as a control. Additionally, 1 of the branded keywords is not only the company name, but also the domain name as well. Also, 1 of the branded keywords is only 1 word and the other 2 are 2 words. As for the rest 9 of them are 2 words and 2 of them are 3 words. So the breakdown of the keywords is as follows:
Google Analytics vs New External Keyword Tool
The New External Keyword Tool, provides us with more data to do our keyword research. One of the most interesting things this tool provides is a month by month breakdown of the search traffic. The old keyword tool provided percentages based off the highest record month in the last 12 months. Below are the results of that comparison
Google Analytics vs. Old External Keyword Tool
As mentioned above, the Old External Keyword Tool, does provide month by month search traffic, but it is represented as percentages. I was able to do the math backwards to get each months search volume since it gives both the percentage and the exact for the most recent month. Below are the results for of that comparison.
Google Webmaster Tools
Google recently updated Webmaster Tools to give us more info in regards to keywords. Most importantly, it gave us additional insight into the number of impressions, clicks, and click through rate, for each keyword. Since this was a recent update we do not yet have months and months worth of data. However, we do have 1 complete month of data, so we can use that as a baseline.
Comparisons of the Keyword Various Sources
Now we get to start comparing the various tools against one another. Now that we have the data before us, lets start to see how well they stack up to each other and, more importantly, which one gets us closest to reality.
New External Keyword Tool vs. Old External Keyword Tool
Comparing these two ‘different’ tools against each other, we see that the differences monthly are not that great. They look to be almost identical. The most recent month, April, is 100% the same. Additionally, the differences between the other months seem to be caused by differences in rounding. This seems to me that the data is being drawn from the same place for the different tools and that it is just the interface that is slightly
different.
Impressions: Google WMC vs. Google External Keyword Tool (New and Old)
This is probably one of the more interesting comparisons. If we take a look at what the impressions are from WMC and the number of searches that the External Keyword Tool indicates we get the following results. Most of the results show that there are far fewer impressions than searches according to Google’s External Keyword Tool. Upon further evaluation, 2 out of the 3 keywords which show more impressions than searches were from branded keywords. The other keyword is from a client with a strong relationship to the keyword. This is can be noted by Google Insights, with the Client Name + Keyword is both a rising and break out search. This one metric, deserves more evaluation, but I think, it may hold some insights into proper and effective keyword research. It is important to note that impressions do not equal searches. Impressions are to Search Volume as Page Views are to Visits. You can get many more impressions then visits, since people will, and do, go on to the next search engine result page. However, since the impressions only reflect Google Search and not its content network, it is not an apples to apples comparison.
Click Thru Rate: Google WMC Vs. Google External Keyword Tool (New and Old)
If we look at the raw differences between the two click thru rates they simply do not stack up. That is the click thru rate from WMC compared against the number of visits to the number of estimated search volume according to External Keyword Tool. But, if we compare the number of visits according to Google Analytics against the number of Impressions from WMC, we get a lot closer. Still not 100%, but there does seem to be a stronger correlation. To me, this seems to indicate that the number of visits in Google Analytics as compared to the number of monthly searches according to Google’s External Keyword Tool, match up quite well. I would not hang my hat on it just yet however. Additionally, we get a lot closer to the fabled ~42.00% click thru rate for these as well.
Visits: Google WMC vs. Google Analytics
Lastly, we can compare the number of visits according to Google Analytics and the number of clicks according to WMC. From the chart below, we can see that they are, for the most part pretty close in hard number differences, but some the percentage differences are quite large. In almost every case we received more visits than clicks. At the extreme, 1 keyword, received 213.70% more visits than clicks. That keyword was a brand related keyword and it was the keyword that was also the domain name. On the flip side of that, the singular version of that received 50.59% less visits than clicks.
Conclusions:
There are a few takeaways we can draw from these comparisons. The first being that the Old Keyword Tool and the New Keyword Tool, seem to be very close to one another. While the data is 100% the same, it is still close enough. Secondly, the impression data coming from Google Webmaster Central seems to be a good number to use if available. While I would not hang my hat on it, it does seem to provide a closer to reality number. Third, when it comes to brand related keywords, things seem to get flipped on their heads a little bit. Last, these numbers deserve more statistical analysis in order to come to more eye opening realizations.
Note:
While the numbers and data that I provided in this post are accurate to the best of my knowledge, please feel free to check my math and/or do some additional analysis. Also, all of the External Keyword Data was extracted as exact match.
