This is how the Google Knowledge Graph works
For many search queries, Google provides additional information above or next to the organic results. These practical boxes are known as the “ Google Knowledge Graph ” and help users to get the information they want even faster. But how does the Knowledge Graph work and how can you use it for yourself? We’ll tell you.
What is the Google Knowledge Graph ?
The Google Knowledge Graph is part of the universal and extended search functions of Google and enriches the organic search results with additional information. In small boxes above or next to the results, data from various sources are brought together and presented.
What content can be found in the Google Knowledge Box?
The knowledge graph is played when Google recognizes a so-called semantic entity behind a search query (see below). The boxes can contain data on people, places, or companies, but they can also display images and entire text excerpts as well as structured data with details on the respective search query.
The prerequisite for this is, of course, that structured data is available. That may be the reason that in the past Google webmasters repeatedly urged users to include structured data on their websites.
Depending on the search term or question, the Google Knowledge Graph is triggered by various factors. Frequently, for example, questions about people (Who is the Federal President of Germany? What is the name of Joe Biden’s wife?) Lead to a knowledge box. Also, dates of birth, definitions, recipes, units of measurement, or distances are often made available by Google directly in the search results.
Semantic Entities – the basis of the Knowledge Graph
The term “ entity ” comes from semantics. An entity consists of a so-called identifier and certain properties. Take Angela Merkel as an example. She is Federal Chancellor and physicist.
The name Angela Merkel is the identifier and, together with the associated properties, forms an entity.
Individual entities, in turn, are related to one another. This context in which entities are embedded is known as an ontology.
Google works with entities to better understand and classify search queries.
Thanks to this model, the search engine can, among other things
- better recognize the relationships between individual words and sentences ,
- in the case of multiple meanings, better understand the intention behind a search query,
- Interpret documents and
- Find relevant entities on a topic .
Knowledge Graph Embedding: How Google understands how to create relationships
What looks like a simple info box to users is a complex network of topics, properties, and relationships in the background. If you wanted to record the interrelationships within a knowledge graph, you would contain a widely ramified network of information. Making this data recordable is known as knowledge graph embedding.
Let’s take a closer look: The main elements of every graph are the so-called nodes. These are the entities. Edges that represent the relationship between the entities run between the nodes. In addition, properties or attributes are assigned to each node, which can also be related to one another.
This can be seen well using the example of films. Every movie is an entity. Google knows that films have a genre, a length, actors, a director, content, a trailer, a release date, possibly a book template, and many more properties.
Its aim is to present this as comprehensively and clearly as possible. On the other hand, it must not be too complex for the users – after all, some properties of the film can again form their own entity with a number of attributes.
In addition to all the properties already mentioned above, the Google Knowledge Box also provides information on the LSP, links to streaming platforms, and ratings. This gives users all the essential information at a glance.
If you are interested in the actors, the book, or the second part of the film, you can click to get to the relevant entity and another knowledge box. In this way, they can move from node to node based on the edges.
Find entities using the Knowledge Graph Search API
If you want to find specific entities, you can use the Knowledge Graph Search API to do so. It uses the standards of the schema.org website, which has set itself the task of collecting, maintaining, and disseminating schemes for structured data.
Various types of entities are relevant for SEO, such as
- Educational institutions
- Music groups
The Knowledge Graph Search API offers you a ranking of the most important entities that meet certain criteria and enables you to organize content accordingly.
Where does the information for the Google Knowledge Graph come from?
In the past, the information often came from the open semantic knowledge database Freebase. Today, Google obtains them through structured and licensed data from websites that the system considers trustworthy, as well as from Wikipedia and Wikidata.
Google uses the last two as evidence of the relevance of an entity. Links within Wikipedia also help Google understand relationships between entities and form ontologies. Topics that have no entry in Wikidata or Wikipedia are usually ignored by the Knowledge Graph.
This is how you get an entry in the Knowledge Graph
You can increase your chances of getting an entry in the Knowledge Graph by using structured data on your website. To do this, provide your content with a markup language. Make sure that all data is correct and up-to-date. Uniformity is also an important point. For example, the information on your page should match the information in your Google My Business profile.
However, there is no guarantee that you will end up in the Knowledge Graph. What exactly is displayed is determined by its own algorithm. Google itself has not yet released any further information on this.
The Google Knowledge Graph has both advantages and disadvantages for SEO
Through the targeted use of semantic entities, Google can now understand search queries much better and place them in a context than a few years ago. This enables the search engine to answer queries better and faster.
This of course offers a great advantage for users: If they get the information they were looking for in the Knowledge Graph, they can save themselves the click of a website.
For websites with high information content, however, this can be associated with a loss of traffic – and this, in turn, can have an impact on advertising revenue and customer acquisition.
On the other hand, there are opportunities for you when your information appears in the Knowledge Graph. Often this is linked to a link to your site. That creates visibility for you. And if the data from the infobox is not enough for users, you can access your website via the link. Against this background, the possibility of appearing in the Knowledge Graph is generally to be rated positively.