RankBrain: Where do we go from here
Having been in the SEO area for 8 or 9 years now, I am used to change, in reality, one of the reasons I like SEO is that it is in constant flux. RankBrain is another matter though, the change is significant. Google will claim that it makes no difference to what you should do – create great content, but it has impacted how we do SEO and how we report on that to our clients.
This is the really difficult thing to consider. Not only do I have to change what I do and how I do it, I also have to explain all of that to my clients because how I present data to them will also change.
How RankBrain is changing the Search Landscape
Much as I would love to go over the specifics, here is some background reading on RankBrain (by Stone Temple Consulting).
Are Keywords Dead?
The keyword is dead, or is it? Where do we start, is it with the search intent, is the search intent related to what was our keyword? The approach I am considering, rather than researching keywords, is to research questions, dig into those questions and plan the site based on answering the questions relevant to your site / product / service.
If a keyword represents a search query then the keyword never dies, but if I optimise a page around a keyword but don’t answer the search intent represented by that keyword then my page will die. It is still important to understand what keywords are being used, it is how we use that information, that now has to change.
Before, I used to give the content developer a set of keywords to optimise around, now I want to present my content developers with the questions I want them to answer, in addition to the words / synonyms of words I would like them to include, but it is the meaning of the word rather than using multiple literal iterations of the word itself, that is important.
Evolving into the Semantic Web
This means in effect, that a page which we previously would have optimised around a single keyword, can now actually perform for a wider range of search terms, once the information on the page satisfies the search intent of those search terms.
How can this work, let us look at general KW research, you get a long list of keywords, I would like to turn those KWs into questions. Then group those questions into common search intents and each of those groups should be reflected with a piece of content on the site which answers that group of questions.
Does SEO have a Place in the Semantic Web?
Creating great content therefore still requires professional SEO support, if we are going to guide content developers properly then our roles have got more complicated rather than less complicated. Working in a localisation company and developing Global SEO / Content Strategies has also increased in complexity, before we localised keywords, but now we need to look at how people across the world ask questions, what are the differences in questions from region to region – those differences will impact on the content optimisation and also new content development in each region.
Note: these are my initial thoughts on how RankBrain will impact us as SEO Professionals, please comment below to add to this conversation, all positive contributions will be published and acknowledged.
Thanks to Jhustin Hall for his review comments via PM.