I recently read a good article from Ryan Jones (for whom I have great respect) where he writes at some length debunking conspiracy theories as they relate to Google not passing keyword referral data to publishers (“not provided”). While I, too, am not big into the conspiracy theory thing, and more or less agree with Ryan’s main points, there was one comment that he made that sparked my interest:
“[myth] Advertisers only get paid keyword data. They don’t get SEO data either. I’ve heard rumors that Adwords now includes SEO data. It does – but it’s pulling that trended data straight from Google Webmaster tools. Advertisers aren’t being given access to any data that Webmaster Tools users don’t already have.”
The first time I viewed the Paid and Organic report in Adwords indicated that advertisers do indeed have access to better data than non-Adwords users. So I decided to pull some data from a client’s Adwords and GWT account and see exactly what was going on. I chose this particular client because
- Search (both paid and organic) plays a significant role in their business
- I have access to both their GWT and Adwords data.
In order to access the Paid & Organic report, one needs to make sure that their Webmaster Tools and Adwords accounts are linked.
For the purposes of this analysis, I’m going to focus on data for “web searches” for Google Search. The 3 data sources that I’m comparing are.
- Google Webmaster Tools
- Google Adwords Paid and Organic Report
- Google Adwords Keyword Data
The Google Webmaster Tools data is filtered for the Web only. This corresponds the most closely to the data that Google provides in Adwords in the Paid and Organic Report. Although an advertiser’s ad targeting settings may be limited to a particular geographic region (such as the United States), the total number of Organic Impressions will be for all of Google Search.
Adwords Paid and Organic Report
One of the things I that surprised me about this report when I first saw it was that it shows Organic Only data. It makes a lot of sense though, as the purpose of this report is to convince advertisers to bid on terms that they rank for so that they can increase their overall Percent Share of Search (last column, Combined Ad + Organic stats).
Adwords Keyword Data
Google Adwords provides keyword level data for queries both on Google Search as well as across Google’s Search Network (which can include Maps, Images, and a wide variety of other sites such as AOL). Keyword level data is also available for Exact Match Impression Share, or the number of times that the keyword displayed an ad divided by the total number of times that the Exact Query or Close Variant of the keyword was entered on Google Search.
Adwords describes the way that Exact Match works (which defaults to include “close variants”) like this:
How it works: Your ads can show for close variants of your exact match and phrase match keywords. Close variants include misspellings, singular and plural forms, acronyms, stemmings (such as floor and flooring), abbreviations, and accents. We won’t show your ads for synonyms of your exact match and phrase match keywords.
In the above example, to get the true number of times that the exact query [safety signs] was entered is not possible since “close variants” is enabled for this advertiser (again, by default). What this should mean for our data is that the number of impressions for an exact match keyword is in fact higher than the total number of organic impressions for the same keyword if the Impression Share is at or near 100%. However, since this advertiser is only targeting the United States, their total number of Organic Impressions in the Paid & Organic report will sometimes be higher. This will be discussed below.
Keyword Level Comparison:
Looking at the top 6 keywords (sorted by impressions) from Google Webmaster Tools we get the following (keep in mind that Keywords are targeted to US only, whereas Organic Data in the Paid and Organic Report is all of Google Search).
There is clearly a difference with regards to the level of granularity to the data provided to Adwords vs. data available in the GWT interface. While it should be no surprise to users of GWT that the data there is healthily rounded, it may surprise some to see that the “same data” in Adwords does not appear rounded at all.
I do not believe that Google is trying to provide better SEO data to Adwords users as a benefit or “thank you” for using Adwords. Rather, Google has a very clear desire to drive more people to spend more on Adwords by making marketers believe they are missing out on a piece of the pie. For example, notice how the Paid & Organic Report oftentimes has significantly more Organic Queries for a keyword even when the keyword has a greater than 99% Impression Share. As mentioned above, that is because the Paid & Organic report uses worldwide Google Search data, whereas advertisers can geo-target their ads. However, finding that information is somewhat difficult in Adwords. Mousing over the question mark does not reveal that information; the advertiser needs to dig into a much longer article to uncover that fact.
As noted, the goal of this report is to entice advertisers to increase their spend to capture a great total Percent Share of Search. Every additional impression is supposed to look like an additional opportunity in the eyes of the advertiser. So, even though the advertiser has near 100% visibility for their target location, the Paid & Organic report will indicate missed opportunity.
Another confusing aspect of these different reports is the fact that the number of impressions and clicks for exact match keywords (as “paid for” in the Keywords Report) differs from the number of impressions and clicks in the Paid & Organic report. In some cases, like [safety signs], there were more paid keyword impressions than paid query impressions. In other cases, like [no smoking sign], there were significantly less paid keyword impressions compared to then number of paid query impressions.
This was likely caused by the fact that other keywords were triggered by the query [no smoking sign]. That being the case, it makes it very difficult to understand how the Exact Match Impression Share metric is actionable for such a keyword. Said simply, how does one truly know what their Impression Share & Percent Share of Search for their target location?
Lastly, I want to point out that while the Paid & Organic report mainly makes an argument for why to advertisers should invest more into Adwords, it still can provide insight as to what extent marketers should invest in SEO. If, and only if, advertisers are using Adwords, they’ll be able to see which paid terms convert well on their site. They’ll then be able to use granular Search Query Volume data (albeit global) for their keywords and this can inform decisions regarding where to put SEO efforts (or rounded numbers from the Adwords keyword tool).
Unfortunately, even though Google vigorously toots the horn regarding the importance of landing page experience for Adwords, SEOs now need to jump through hoops to determine the relationship of their keywords to landing page experience. (This is not a post about analysis techniques in a “not provided” world, so I won’t go into detail; more than enough ink has already been spilled on that issue). For what it’s worth, Google does make it difficult to see how much money may actually be getting wasted on Adwords by not matching up Adwords query data with cost data in Google Analytics.
Google now shares granular Webmaster Tools sourced data with Google Adwords in a way that differs significantly from the data provided outside of Webmaster Tools. This is not intended to get SEOs to begin advertising, because they can not get granular keyword data in Adwords.
However, the purpose of the Paid & Organic report is indeed to promote Adwords spend by making marketers believe they have room to improve their Percent Share of Search (which, in many cases, is altogether true). The actual data sources that are available are confusing and can seem contradictory, which also probably leads to greater Adwords spend. I am not of the opinion that all of Google’s decisions regarding the data they share is carefully thought out down to the last “crossed t and dotted i”; but as an outsider it does appear that some of their data sharing practices are suspect.
Lastly, Google presents a highly hypocritical message with regards to the importance of landing page relevancy by only sharing keyword data to analytics tools for paid search but not for organic search.