Rank Ranger Blog

What are Google Entities?

What are Google entities

When I started my journey into semantic SEO, I discovered that there are many concepts that I needed to understand before getting started.

But, I also found that many of the resources online were painfully difficult to understand. I often read a sentence more than once and came out with either a hazy understanding or worse yet, more than one possible explanation.

Believe me, nothing will confuse you more than having more than one way to interpret a sentence.


via GIPHY

More often than not, the reason for this confusion was I didn’t have the necessary vocabulary to pinpoint what the writer meant.

So, after working hard to understand semantic SEO, I decided to create content that spells everything out in simple English.

Also…

This post is a deviation from my usual content, as I generally try to give practical tips and strategies in my blog posts. But, since I was confused, I assume others are too. This is my attempt at giving you a clear understanding of some important concepts.

And, that brings us to the subject of this post.

If you were wondering what a Google entity really is, this post is for you.

In it, I hope to clarify:
  • What an entity is
  • What the various component parts are
  • What the Topic Layer is

Once you understand these points, not only will your semantic SEO geek friends love you, but you’ll have both the knowledge and vocabulary to research the topic on your own. And since semantic SEO is not just a few tactics and best practices, having a real understanding can really help you get more traffic.


via GIPHY

So, as we dive into the topic, let’s first understand why entities are important to search engines.





What are the Benefits of Google Entities? 



Entities are important because they help search engines categorize information in a way that helps them:
  • Understand user queries
  • Answer these questions

So, to understand this, put yourself in the shoes of a search engine. Search engines are designed to solve a problem. How do you organize information on the web in a way that’s easily searchable by its users? How do you bring the right content to answer a searcher’s queries?

In order to better solve these problems, search engines moved over from lexical search engines to semantic search engines.

Lexical search is where the search engine finds keywords in search queries and matches them with those keywords in web content.

Semantic search on the other hand is an attempt to improve the accuracy of search results by attempting to understand the search intent through understanding the meaning of the words in the search query and matching it with content that truly answers the search query.

(In reality, Google uses both approaches.)

In a semantic search engine, an entity is a meaningful unit for organizing information. The reason for this is in order for Google to ‘understand’ the search query and match it up with the right content, Google needs to understand what entity the search is about and how it relates to other entities. Unfortunately, I won’t be able to explain how that works in this post but take a look at the semantic triples section in my blog post on semantic search.



What Are Google Entities? 



Entities as defined by Google are "A thing or concept that is singular, unique, well-defined, and distinguishable.” These entities are often characterized by name, type, attributes, and relationship to other entities.

Now with that definition in hand, you might be wondering, surely an entity refers to an object in the real world? How do entities include concepts?

If you asked that question, I’d be inclined to agree with you, we generally associate the word ‘entity’ with things in the real world.

But, as I mentioned before, the goal of Google entities is to organize information, and entities are meaningful units for organizing information. From the perspective of organizing information, there is no real difference between abstract nouns (such as concepts, colors, and feelings) and concrete nouns (or real-world things).

Google, therefore, treats them in the same way as it treats real-world entities.


The Anatomy of a Google Entity 



If you were to look at a Google entity through a (metaphorical) microscope, what would you see?


via GIPHY

Entities are made up of different parts and understanding each part will help you to see how they relate to one another. Understanding their relationship will help you create content that Google ‘understands’ and doing that could help you to rank higher and get more traffic.

The first thing to understand is that entities exist in a catalog called a Knowledge Graph.



What is the Google Knowledge Graph? 



Google’s Knowledge Graph is a repository of information used to store entities together with their descriptions and attributes. What’s more, the Knowledge Graph also structures the entities in a way that makes connections between entities and concepts. Understanding this structure is a key concept in understanding Google entities (more on this later).

Google's Knowledge Graph gets all of this entity information from resources such as:
  • The CIA World Factbook
  • Wikipedia
  • Wikidata

Google also gets entity information from the open web.

This means Google will get entity definitions, categories, entity types, entity properties, and more from these sources. Google then stores all of that information in structured form in the Knowledge Graph.

Now that you are familiar with the environment in which entities exist, let's examine the entities themselves.



Entities and Their Properties 



As I mentioned above, entities are made up of component parts. Let’s examine these parts separately:


1. Unique Identifier 


Simply put, an entity knowledge base needs to have a way to identify each separate entity. This means each entity needs its own unique identity.

To see a unique identifier in the wild, let’s take a look at Google’s Knowledge Graph API explorer.

In the screenshot below, I’ve typed the entity, John Lennon, into the query field.


Google's Knowledge Graph explorer with 'John Lennon' in the query field


If you scroll down you’ll see "@id": "kg:/m/01vsl3_",


The Knowlege Graph explorer showing the entity ID


This is the unique identifier. You’ll notice that the unique identifier is not the entity’s name. It’s an identifier that’s unique and machine-readable.

Now you might be wondering why Google doesn’t just use the entity’s name. The reason is many entities share the same names. The identifier, on the other hand, is unique to each entity.


2. Entity Name

Each entity has a name. As I mentioned above, unlike unique identifiers, entities may share names. For instance, Taj Mahal is both a historical building and a musician.


Google SERP for the term 'Taj Mahal'


What’s more, some entities might be known by more than one name. For example, typing Louis Armstrong or Satchmo into Google will bring almost identical results.

These alternative names are considered aliases.

Now, you might be wondering…

If each entity has a unique identifier, why do they need names?

The entity name is important because, even though the unique identifier is a way for the search engine to identify the entity, the name is how an actual person will identify the entity. People will include the name of the entity in search queries as well as in content.

Google on the other hand will see the name in a search query or in a piece of content and based on context attempt to match the name with the entity in the Knowledge Base.

As you can see in the screenshot below, the entity name is John Lennon (in case you hadn’t guessed. 😉)


The nowledge Graph API explorer showing the entity name


If you want to see the entity name in the wild, type it into Google and if Google is confident it knows the entity, Google will display the name at the top of the entity’s Knowledge Panel.


Knowledge Panel for the entity John Lennon showing the entity name


3. Entity Types

Entities are grouped into entity types. Entity types are simply categories that group entities with similar properties together.

Typical entity types are people, places, things, or movies.

So, looking back in Google’s Knowledge Graph API explorer, you can see the entity type by looking for "@type”. (Apparently, to Google, John Lennon is both a thing and a person.)


The Knowledge Graph API showing the entity type


4. Entity Attributes

Entities are generally characterized by sets of attributes. Each entity type will typically have its own sets of attributes.

So for instance, entities categorized as people will typically have attributes such as date of birth, family members, height, etc.

Places on the other hand might have latitude, longitude, country, postal code, etc. as attributes.

Sometimes these attributes are entities in their own right. For instance, country. When this happens, these attributes are not treated as attributes, but as related entities.

To see attributes in the wild, let’s take a look at John Lennon’s Knowledge Panel.


Knowledge Panel showing entity attributes


As you can see, the Knowledge Panel tells you:
  • When and where he was born
  • When and where he was assassinated
  • His spouses
  • His children

The words in blue are attributes that are also entities in their own right. If you click on any of them Google will redirect you to that entity’s SERP.


5. Related Entities

Related entities describe how entities relate to each other. For example, to Google, both John Lennon and the Beatles are entities that are associated with one another. The relationship is John Lennon is a member of the Beatles.

To see the relationship for yourself, type ‘the Beatles members’ into Google.


Google SERP showing entity relationships


In the screenshot above you can see all the entities that Google considers to be members of the entity the Beatles.

Google uses these entity relationships to answer user-generated questions. So for instance, in the example above, by typing ‘beatles members’ into Google, Google understands that the user is in essence asking ‘Who are the members of the Beatles?’.

Google already ‘understands’ the relationship between the Beatles and its members and is able to surface that information in the SERP feature above.

Understanding how entities relate to each other is super powerful information when it comes to SEO. By including related entities in your entity content, you’ll be building context for Google to understand what entity your content is about.



How Entities Have Changed Search 



By using entities as a unit for categorizing information, and by organizing entities into relationships, Google is able to create a rich search experience. This means when you search Google, you often see SERP features that include related entities and entity attributes…

What’s more, this has given Google the ability to bring short answers to users’ questions. You can often see these answers featured in SERP features.

As a side point, since Google presents entity information in SERP features, tracking SERP features is crucial for your SEO.

For instance, if I type the word ‘weather’ into Google, Google is able to figure out that I’m requesting information about the entity ‘weather’.

Google then looks for the answer to this question by looking at sites in its index. Google then compiles the information in a SERP feature.


Weather SERP feature


Now as it is, this is pretty impressive, but, Google has taken this understanding much further by creating the topic layer.



What is the Topic Layer? 



In 2018, Google added what they called the Topic Layer to the Knowledge Graph.

Now, as I mentioned above, Google was already able to connect entities and bring them into the SERPs in SERP features.

So what did Google add by creating the Topic Layer?

The Topic Layer was designed to help Google users by looking at search as a journey rather than only a way of getting quick information. In other words, there are topics that are too complex to take in quickly. These topics might take many searches over multiple days.

The topic layer was designed to help users explore a topic by providing ways to keep track of their searches and by suggesting things to explore next.

This is a new paradigm in how Google helps its users. Google was great at answering questions. Google can now help users along greater search journeys.

In order to do this, Google understands that by arranging content into a topic/subtopic hierarchy.

To see this for yourself, take a look at the results page for the search term ‘marketing’.


Google SERP for the term 'marketing' showing subtopics


As you can see in the Knowledge Panel, Google presents subtopics such as:
  • Purpose
  • Goals
  • Kpi

These are all subtopics within the general topic of marketing. What’s more, Google also suggests other related searches including:
  • Internet marketing
  • Marketing strategy
  • Sales
  • Social media marketing

Again, these are all ways to arrange Google’s entities into topic/subtopic hierarchies.

So, if marketing is the general topic, then internet marketing, marketing strategy, sales, and social media marketing are all subtopic entities.





Understanding Entities - Where to Go From Here 



Hopefully, this post has shed some light on the topic of entities in search. Although I have not included any practical tips or strategies, I hope you will have a newfound understanding of the topic which will give you the knowledge to research and understand what to do next.

The way I see it, knowledge is power if you figure out how to use that knowledge. So if you want to take your knowledge of semantic SEO strategy to a place where you can use it to create a semantic SEO strategy, the next step is understanding the difference between keywords and topics.

About The Author
Darrell is a content marketer at Rank Ranger. While working as the SEO manager at a small marketing agency, Darrell discovered his love of marketing and SEO.


Start your free trial


Get the ultimate SEO tools with Rank Ranger
Start Free Trial
No Credit Card Required