As early as 2001, Tim Berners-Lee, the ‘inventor’ of the Internet, predicted the coming of the semantic web. With Google’s Hummingbird update and its increasing focus on entity search, it seems like we’re not too far away from that coming true. This article will introduce to the concept of entity search (for the purposes of this article, entity search and semantic search will be used interchangeably).
First, let’s quickly define semantic search.
According to Tamas Doszkocs of WebLib:
“Semantic search is a search or a question or an action that produces meaningful results, even when the retrieved items contain none of the query terms, or the search involves no query text at all.”
For example, try searching for ‘that movie that’s backwards and the guy can’t remember anything‘. Google will correctly return the results for the movie Memento.
Now, that really turns traditional keyword-focused SEO on its head; simply because there are no keywords to target.
So how does Google know how to return the correct result?
It uses Entities.
What is an Entity?
An entity could be a person, a place or a thing. Okay, that seems kind of obvious. It’s easy for a living, breathing human being to understand that concept but not for a machine. Especially a machine that has trillions of web pages in its index.
Entity search is Google’s attempt to better understand user intent with regards to search queries.
It might be easier to explain this with an example.
Let’s look at what happens when you query, ‘when was martin luther king jr born’.
Firstly, you can see that it very conveniently returns a date right in the SERPs so you don’t have to visit a website to find an answer. Secondly, you can see that his date of birth is in bold in the organic search results.
In order for Google to do this, it first has to understand that:
- Martin Luther King Jr is an entity, specifically a person
- You are looking for a particular attribute of that entity, specifically a date
Google has a large database of entities and knows how they are related.
For example, it knows that Martin Luther King Jr is also known as:
- martin luther king
- reverend king
- dr king
Also, that alternatives to the word ‘born’ are:
- date of birth
- date born
This means that Google knows what the user is looking for even if you use any of the following searches:
- when is reverend king date of birth
- when dr king dob
- mlk was born when
This is not because Google searches for keywords and return pages that it thinks are relevant.
Instead, it is able to return the answer because it recognizes Martin Luther King Jr as an entity (person), and that the user wants to find out an attribute (date of birth) of that person.
Google is able to do this because of its Knowledge Graph. This Knowledge Graph attempts to understand “real-world entities and their relationships to one another: things, not strings”.
Google seems to be laying out the groundwork to more advanced entity-based searches with the release of the Hummingbird update. I’m quite sure that this will be a very important consideration to SEO in the very near future, but for now, keyword-based searches are still worth focusing your SEO efforts on.