This is the SEO’s quest, to understand the mythical “Algorithm”, the heart and soul of Google’s search results. How does Google decide what to rank? Why this page over that page? Why images here instead of text? Why text instead of video? Why does this search result get Knowledge Graph results, but this other one doesn’t?
Decoding what Google is thinking, why they choose the results they do, is a critical part of the job of an SEO. An SEO needs to learn to think like Google.
So what do we know about this mythical algorithm? That depends on what you mean by “know”. Google doesn’t usually just come out and say what is and isn’t a ranking factor. They do drop hints, sometimes, and they often drop misinformation as well. But some things they’ve stated explicitly, which you can find by keeping up with the Google Webmaster Central Blog, and in Google’s quality rater handbook. Add to that all the data from SEOs who have been poking, prodding, and testing Google since the day they launched (which you can find on sites like Moz, SEMrush, SEJ, etc.). It’s fair to say that we know quite a lot.
These charts are from Moz, a leader in the SEO community and one of the oldest and best resources for SEOs. These charts are a few years old, and the exact percentages likely aren’t accurate, but they still do a fantastic job of breaking down the various ranking factor buckets that are still very much relevant today. This data is gathered by polling leaders in the SEO community and aggregating the data, and tends to be pretty spot on. We have a really good idea of what Google’s algorithm takes into account. We also know quite a bit about the relative weighting of each element…or at least, we did.
Enter Google’s RankBrain. Here’s another useful bit of info from Moz, from a Whiteboard Friday. RankBrain, in summary, is a machine learning system designed to adjust the weighting of various ranking factors based on the search query, in order to provide better results. It could apply different ranking factors to different queries, with different weights to each factor, to provide the best possible contextual result. This is huge.
It means that there may no longer be an “Algorithm” per se, with fixed weighting for each factor. The factors still exist; RankBrain is using inputs to make these decisions, using all of the data that Google has gathered. But the list of factors and weighting is constantly shifting, varying from space to space and query to query.
SEOs have always been trying to hit a moving target, but now it’s multiple moving targets, blindfolded, with one arm tied behind their backs. Damn.
Why does this matter? Because it changes everything. A lot of the old stuff still matters. You still have to optimize most things in the same basic way (and this course will cover all of those things). But now it’s more important than ever in the SEO world to carefully understand your target audience, your competitors, and what Google is showing for various queries in your space. You have to reverse engineer, space-by-space, what is and isn’t working, what is and isn’t important SEO-wise. There are rules…but there are now all sorts of exceptions to those rules.
And so, the job of the SEO becomes one not only of knowing the rules, but of knowing the exceptions, and how to find them. And that starts with knowing the options.