If you’re doing SEO or if we do SEO for you as a client, you know that there are hundreds of signals that help determine how a site ranks. We already know
that for local searches, Google tries to return the most relevant local queries based on where (location) you’re searching from.
But how does Google do this? With an algorithm, of course. The difference today, though, is that Google is now using more and more Artificial Intelligence
to figure out how to rank you.
The AI that Google uses is lovingly named RankBrain. It uses machine learning to better understand the type of search you’re doing and tries to find exactly
what you may be looking for.
RankBrain uses artificial intelligence to embed vast amount of language into mathematical entities called vectors. If RankBrain sees a word or phrase it
isn’t familiar with, the machine can make a guess as to what words or phrases might have a similar meaning and filter the result accordingly, making
it more effective at handling never-before-seen search queries (source: Wikipedia).
Let’s look at the 3 main types of search:
- Informational (research)
- Navigational (typing a company name into the search bar to visit a pre-determined website)
- Transactional (e-commerce, sale, downloads, etc.)
Now, everyone searches a bit differently based on experiences, cultures, and more. But, RankBrain tries to think more like a human and return the most
relevant results. It uses an artificial way of thinking and learns from past searches and results (most likely based on whether a person re-searches
and/or click data from a set of results for similar queries).
So while the typical (old) algorithm used math and science to return results, the same is still true — however, RankBrain is a machine intelligence
that learns and gets better at figuring out what it is your intended search should find.
And while not too much has changed in term of results, over time they just might.
Intelligence Shaping the Algorithm
If we assume the signals are the same, the difference here would be that machine intelligence is able to adjust the ‘weights’ of those signals on the fly
— in order to return better, more targeted, more specific results.
Optimizing for RankBrain
At the moment, the best thing we (or you) can do is to continue optimizing those signals we already know of and keep an eye on results. As RankBrain improves
and becomes a bigger part of the Google Algorithm, we’ll have to adjust our strategies and tactics in order to better signal the Artificial Intelligence
for intended searches.
Written by Joe Schaefer, SEO Specialist at Brawn Media.