You might be hearing a lot about how machine learning or "ML" is influencing how content creators think about SEO.
Machine learning — the ability for computers to meaningfully interpret large volumes of data without being explicitly told exactly how to do so by a programmer — is changing a lot of the conversation around content creation on the web.
If It Walks Like a Duck and Quacks Like a Duck...
Remember how your 4th grade science teacher told you that “computers can only do exactly what they’re told to do”? Machine learning calls that particular bit of wisdom into question.
Imagine you are given 10 pictures of different animals, and you are asked to identify which one is a duck. This is the kind of problem that’s trivially easy for a human, but historically quite difficult for a computer because, in traditional programming, you have to explain in detail exactly what the computer is supposed to do. So how do you tell it to “find the picture of the duck” when the computer doesn’t know anything about ducks, feathers, bills, webbed feet, or anything else that we think of when we think about ducks?
That's where machine learning comes in. In its simplest terms, machine learning lets us “train” the computer by having it analyze lots of pictures of ducks, along with lots of pictures that are not ducks. At first, we tell it which pictures are which, and the computer “learns” to recognize patterns in the duck pictures, and gets very good at guessing whether the picture is a duck or not. What makes the ML algorithms even more useful is the fact that, once trained, the computer can recognize pictures of ducks that it has never seen before.
SEO Implications for Content Creators
Enough about ducks. What does all of this really mean for those of you on the ground — content creators who don’t have a degree in computer science and won’t be crafting your own machine learning algorithms any time soon? The answer can be distilled down into one simple fact: Many fundamental aspects of SEO are changing.
I’m surprised when I see SEO conversations still revolving exclusively around page titles, meta tags, and keywords. While those facets of SEO are still relevant, and I would never recommend implementing an SEO strategy that wholly ignored them, the SEO conversation needs to include user experience. Google and other search engines are now “training” their algorithms on how to identify users who are getting what they want from a website. Read more from Neil Patel and Moz about how Google is likely using behavioral data to affect search ranking.
Ultimately, instead of thinking about keywords and meta tags, you need the user behavior of your website to show indicators that people are engaged in the content and taking relevant actions. In other words, you want Google to come up with “yes” when it asks the question, ”Does the behavior of users on this website appear to correlate with the behavior of users who are getting what they want out of a website?”
Staying Ahead of the ML Curve
How can you keep your website in front of advancements in machine learning? The answer here is simple and not entirely surprising: Keep your focus on creating quality content that engages users.
How can you use advancements in ML to your advantage? Spoiler alert: It all comes back to good content.
The old-school approach to SEO involved crafting very wordy, text-heavy pages that were stuffed with keywords and phrases you wanted to rank for. You were writing content “for search engines.” Now, thanks to advancements including machine learning, search engines are a lot smarter than they used to be, and they’re no longer interested in whether the content is good for them. Instead, they want to know whether it’s working well for an actual human.
So you'll want to think about the experience users will have, rather than the keywords the search engine will see. For example, rather than listing your service offerings in a bulleted format, you might give users a visual grid of services that expands to show more information when clicked. Likewise, instead of asking users to work through your menus to find out which content is relevant to them, provide a friendly interface to guide them to the right place (by asking questions about who they are, what they’re looking for, and so on).
In the end, leveraging ML with useful and compelling content that encourages users to regularly browse, click, and interact falls into the "win" category for all: your organization, your brand, your SEO rankings, and your viewers.