Today is a very sad day for SEO professionals. As of today, Google has eliminated all keyword data through encrypted search using HTTPS. The first effects of encrypted search started back in October 2011 when Google started hiding keyword data for anyone that was searching while signed into their Google accounts. Initially they said this would have very low impact on data, however over the last two years we have seen those instances increase exponentially, sometimes accounting for as much as 75% of organic search data. Well now those days seem like a world full of rainbows and butterflies. I’d take 25% any day over 0%.
What are the impacts of losing keyword data?
So what does this means for those of us in the SEO field, or any businesses that cares about their presence in organic search? The first, and in my opinion, highest priority implication is on keyword strategy, or how we determine which keywords to target across our website and content strategies. Of course we will still have search volume estimations so we will know approximately how popular a keyword is, but we’ll no longer have insight into how that keyword performs for a specific website or brand. If all you care about is driving more traffic, this might not be very important, but for those that focus on driving the right kind of traffic knowing if a keyword actually drives conversions is vital information.
“Why not use paid search data?” you might ask. We most certainly do use paid data and always have, however it’s not always the apples to apples comparison we need. For example, there may be a keyword group not included in the paid campaign due to high CPC’s, or maybe it’s just not a focus for the paid campaign. It may still be a great keyword group for organic, but without any conversion data it will be impossible to confirm. However this change will make utilizing paid search data that much more important.
Reporting and over-all performance analysis will also be impacted by this change. Our view of successful SEO is based around the increase in NON-BRANDED organic search. For the majority of the clients we work with, they are already ranking in #1 placements for their brand terms, so we make sure to break that traffic out so as to not claim credit for that. Without keyword data, there is no way to determine branded from non-branded, so it all now rolls up under one big ambiguous organic umbrella. Also, any type of individual keyword or keyword group analysis is basically out the window. There has been so much talk about moving away from using ranking to communicate results, since they are so unreliable in terms of the source. However with the loss of all Google keyword data in analytics, ranking reports are looking pretty good about now.
What are the alternatives?
As already mentioned, paid search data is going to be one important source of information. Google Webmaster Tools is another, as it does provide search query data in the form of impressions, clicks, CTR and average position. Unfortunately, still no conversion data. Most likely, and considering I’ve only known about this for less than a day this isn’t entirely flushed out, the best analysis is going to be a deeper dive of landing pages. If you have an understanding of what keywords each landing page is ranking for (this is where the Webmaster Tools report comes in) then you can draw some conclusions based on the conversion rate of organic search to landing page. It’s not even close to perfect and allows for a lot of mistakes with assumption, but at least it’s something more than just looking at over-all organic traffic to the site.
Another possible area for analysis is focusing on other search engines, such as Bing, for your keyword analysis. That is of course dependent on getting enough traffic and conversions through these other search engines to provide actionable data. And even still, that data needs to be taken for what it is, considering that Bing users are typically considered a slightly different audience to that of Google.
As we do more testing here, I’ll provide an update as to how we are coming to terms with this change. If you have ideas or insights in this area, please comment, we’d love to hear what others are coming up with.