Natural Language Processing has been revolutionary. It’s transformed the way Google analyses and understands the text on a web page, and while keywords are still an integral part of SEO, machine learning has come a long way from the early days of basic word counting. Algorithms now teach themselves how to better understand the user, what you’re searching for and ultimately, what the best results will be for your search query. It’s all about user experience.
What is Natural Language Processing?
To keep things simple, Natural Language Processing is a collection of techniques which are designed to interpret human language by breaking it down into shorter chunks. This allows Google to understand the relationships between words and how those words work together to create meaning. This means that the search results are better matched to the search query’s intent, giving the person searching a better experience.
If you’d like to see for yourself how Google’s Natural Language API works, you can give it a go via their Cloud Natural Language system.
You’ll see that Google identifies entities such as things, people and places as well as how important they are within the text (salience). It will then make associations between the words and effectively labels them in various ways to determine what their function is in the sentence and how important they are in determining the meaning of the content.
How does this affect content writing for websites?
Essentially, it’s changed the game. Where once upon a time it was encouraged to write more and include certain keywords a number of times, NLP has made clarity and readability take precedence. The more complex our sentences, the harder it is for the Natural Language algorithms to understand. Over-complication of sentences means we’re likely to deviate from the main purpose of the text, confusing the algorithms. Less certainty in your content writing, means lower rankings.
Practical Tips When Writing for Google’s NLP
Connect Questions to Answers – Stick to writing one or two sentence answers, much the same as what you’d expect to hear from a voice assistant.
Be Clear – Make it explicit what it is that you’re referring to. For example, “200 degrees Fahrenheit” is clear, “200 degrees” is not.
Keep Sentence Structures Simple – Use conversational language where suited and be conscious of word association. Convoluted sentence structures will not only confuse the algorithms but readers, too. Tools such as Hemingway can help with this.
Be Specific and Absolute – The NLP algorithms have a hard time understanding “it depends” answers. Answers that are direct and to the point are more likely to be rewarded with better rankings.
Avoid Misinterpretation – If a sentence you’ve written could have more than one meaning or if someone has to reread the sentence more than once, a rewrite is probably required.
Improve Salience – Lose any ambiguity about the subject matter by adding further terms that are associated with the thing you’re focusing on. One simple and effective way to do this is to use Google’s “People also searched for” and “People also ask” sections of the search results to give you a better idea of what other information you could include.
Google rewards pages that follow the searchers journey, this is because it has a lot of data about the sorts of things users will search for next. If you use the information provided by Google to build your own content, it will become a better destination for the user, meaning higher rankings for your page.
So what are the key takeaways?
Avoid the waffle, stick to writing in clear and simple sentences with definitive answers, and try to answer as many questions as you can. Enjoy!