Getting the Data You Need to Make Decisions in a Post-Keyword-Provided World
Marketers and organizations are scrambling now that Google has turned off keyword data from organic visits for all users, including those who aren’t logged in. While bloggers and SEO thought leaders evaluate why Google went this route and what the change really means, r2i’s search, analytics, social and strategy teams share our reactions and focus on the tools and processes we had in place in preparation for this day. Part of the general concern seems to be around the mindset that Google Analytics was the “end all, be all” for keyword data. However, we have never leaned on one singular source for information. Google Analytics keyword data, especially when they introduced “not provided” in 2011, has always been peripheral; we use multiple data sources to confirm what we believe to be the truth about our keyword data.
First, our reactions; then our solutions.
Insights from r2i Search, Analytics and Social Experts:
Jeff Budd, SEO Strategist: This ensures our team will collaborate more to glean the best insights from the data that we do have. The Analytics team will leverage its access to AdWords to mine keyword data for SEO purposes. Teams will most definitely rely on Google Webmaster Tools a lot more as well.
Jorge Castillo, Director, Analytics and ROI: Keywords are kind of 2010. We live in a packaged world of content (which we know) and the sharing that occurs through dark social (which we don’t know).
Chris Montcalmo, SEO Strategist: It was only a matter of time as spammers had been gaming Google Trends for years, hijacking hot keywords with “thin content” posts loaded with ads and spyware. The sooner we get away from the “keyword” mentality and embrace latent semantic indexing (LSI) and co-citation, the better off we’ll be.
Natasha Jarmick, Senior Account Strategist, Digital & Social: We all knew this was coming from Google, however it's a surprise how quickly they pulled the rug out from underneath Google Analytics users. This change will most heavily impact small businesses and independent bloggers/site owners – which is unfortunate, seeing as previously, SMBs were Google's bread & butter. Additionally – this could prove to be an opportunity for other search engines, such as Bing and Yahoo, to become the preferred engine for marketers. Perhaps site owners will begin integrating Bing search for their internal site search engines instead of Google, in order to better align with robust data and SEO analysis.
Kara Alcamo, Senior SEO Strategist: This is not a surprise even though its implications are frustrating at best, especially for small businesses. That being said, there are other ways to continue to get the information you need to measure performance and make decisions. Start by going back to SEO basics with some key questions. What do you do with keyword data? Why do you need keyword data? The most obvious need and use for keyword data is to identify the ones that will work best for conversions. We rely on keyword data to assign value and an optimization priority to specific keywords. With or without Google Analytics, we use several alternative tools and methods to access keyword data.
1. Use Webmaster Tools to Estimate Visits per Keyword
We’ve been looking to Google Webmaster Tools (GWT) more heavily in the past couple of years, and now it looks like it will become an even greater resource. Look to clicks on queries to estimate traffic from a particular keyword, and look at the landing page(s). More clicks in GWT means more visits to the page, with the caveat that the numbers will not match up exactly.
Also look at the proportion of visits per keyword. In Analytics, look at the landing page filtered to include only Google visits, then match up proportions of query visits to the total number of not provided visits during the same time to estimate visits for each keyword. If most are <10, then use a longer date range.
While doing this, keep in mind that GWT data is imprecise, and it should mainly be used to analyze proportions and change rather than hard numbers.
2. Look at other search engines
Google effectively turning off their keyword data is a big blow, but it’s important to remember that they’re not the only search engine. Bing, Yahoo, Ask, Yandex, and others are still providing keyword-level data, giving us insight into the conversion rates and efficacy of particular keywords.
3. Start at landing pages instead of queries
Look at your rank tracking software and in Google Webmaster Tools to see the pages on which visitors are landing and the keywords they could be using to get there. Track conversions from that point instead of starting at the keyword level. The goal then is to determine what groups of keywords give you the greatest return.
4. Test in AdWords
Privacy concerns or no, Google still provides keyword-level data in AdWords. Look to your paid search data for conversion information on particular keywords. Test keywords in AdWords before or while optimizing. In fact, test a variety of keywords and look at conversions and conversion rates.
Many SEOs have already been leveraging AdWords to test certain keywords; will it work, is it relevant, and will people respond? This change from Google just increases the price of this data. We can’t optimize a page and then see what keywords work; we have to get the data from AdWords. You may have previously spent $500 in AdWords to test a specific keyword, and anything more expensive than that would just get tested organically. Now, you may be willing to spend a bit more to understand the traffic and opportunity for a set of keywords. The only caveat with this is that some keywords are just too expensive to test; you will need to rely on a combination of methods to evaluate keywords.
None of the above methods will give you the direct Truth that we were once able to get from Google Analytics keyword data, namely what specific keywords result in the most conversions. But the fact is, we haven’t had Truth with a capital “T” since Not Provided was introduced. Percentages of Not Provided results have been increasing steadily, and we’ve had two years to come up with alternate ways of getting the information we need to make big decisions. Google has cut the cord, and it’s time to transition completely. There may not be direct data anymore, but, there is more than enough indirect data to get us to the same endpoint.
The Truth lies in where they intersect.