If you’ve ever searched for an actor, historical figure or even a lamb stew for a lazy Sunday you’ll have probably also noticed a box on the right hand side of the Google search results. This is the visual output of an engine known as the Google Knowledge Graph, acting to link together separate facts and concepts while improving the relevancy of results for user searches. Unlike Latent Semantic Indexing, the Knowledge Graph deals in wider ideas, expanding its remit beyond simply keywords.
What Does the Knowledge Graph do?
Imagine you were to search for, say the actor Daniel Radcliffe. Not only do you see a collage of images, you also get information from his wikipedia page, his height, his age, his parent’s names, the TV shows he’s been in, all the Harry Potter films, a selection of his most famous quotes and, finally, a list of affiliated people such as family and other cast members. For Google to do this it can’t just simply link keywords together. It needs to accurately collate information from a variety of sources and turn it into a tangible ‘something’. In short, it needs to know just what a Daniel Radcliffe is.
On the face of it this appears a fairly straightforward task, however ask five different people to define Mr. Radcliffe and you might get five different responses. He could be an actor, a person, Harry Potter, a charity ambassador or a twenty-something heartthrob. It all depends on who you ask. Google knows this and tags the search term accordingly, allowing content related to acting, or Harry Potter to appear – just like Hogwarts magic.
What does it mean?
So you might be wondering where the “graph” part of this comes into play. Simply put, it’s a technical term used to describe the way in which sets of objects connect to one another. A link graph shows how web pages are connected through hyperlinks, a social graph shows how people are connected and knowledge graph expands on these by linking together people, places and objects – reporting on these as facts to the user.
How does it work?
To prevent its users from drowning in a flood of useless trivia, the Knowledge Graph only displays facts it has identified to be relevant to the search query. Search for Picasso and a series of his paintings will be brought up, search for Bill Gates and you’ll see his net worth. That’s not to say Bill Gates has never painted, it’s just a reflection that users are generally more interested in his business life than creative endeavours. The concept of “entities” (something that has a value in the real world) is central to this functionality, and as Google becomes smarter it’s become increasingly effective at identifying the most salient of these, bringing up the most relevant information accordingly.
How the Knowledge Graph helps its users
The Knowledge Graph is yet another example of how search engines are looking to expand their remit towards that of a direct answering service, moving away from simply referring people towards the source of the information. For its users this makes searching for individual facts much quicker, as they no longer have to parse sites for isolated snippets of information. On the other side it has raised concerns with the owners of some sites, believing that users will now be less inclined to click through to the original source.
In reality the Knowledge Graph simply redefines the rules of engagement, with companies seeking to embrace it ultimately benefiting the most. To do this pages can incorporate into the system, becoming the established voice of authority in relation to a certain query. For example were you to write a step-by-step guide on how to draw a pigeon, snippets from your site, as well as its link, would pop up within the graph display – driving more traffic towards your site.
Becoming part of the Knowledge Graph
Understandably, becoming established within the Knowledge Graph for target queries can be immensely valuable. However, Google won’t just pick any snippet at random, it needs to first ascertain that your information is the best answer to a question. The most effective way to become a featured snippet is to always keep the type of query being asked at the forefront of your thinking. This allows you to write highly focused content that provides a definitive solution to the problem at hand. Formatted in an easily displayed manner, this approach ensures that the user stays at the heart of the experience; establishing the foundation for any effective SEO campaign. Aside from this there are a couple of actions a site can take to improve its chance of being included.
Google demonstrates a strong preference for sites that display schema mark up, effectively a way of structuring data so that search engines do not have to attempt to ascertain the solution, instead the mark up displays this in a more, well, structured way. Examples such as cooking times in a recipe, opening hours for a shop or the fact a product listing is, in fact, a product listing, not only helps identify what the content of a site is about but allows for the inclusion of featured snippets on search engine results pages (SERPs). These are additional pieces of displayed information that can help in driving traffic through a site, increasing the chances of making the leap onto the graph card.
The Knowledge Graph is constantly updating to provide the most up to date, relevant results. Therefore providing more complete, better-formatted or simply more accurate answers allows a source of information to leapfrog another – even if it may not display as highly in the more traditional organic rankings. Google provides a report feature to flag up false information for review and this is emblematic of the fact it is not quite perfect in its execution. However as it collates more data it will only become increasingly more accurate, the goal to create the most human experience possible. As they say, knowledge is power and it’s safe to say the Knowledge Graph is just the start.