I read a lot of patents, many of which may-or-may-not apply to SEO or may not be applied at all by Google. Their recently granted patent ‘Related Entities’ clearly is not one of these. It is currently being applied and more, it gives us significant insight into their systems beyond what’s covered directly.
In fact, where I intended on writing this article as an analysis of the patent itself, that will only be a means to an end. We will look at some details as it is interesting and an understanding of it is needed to grasp the connections to SEO at large but it’s the “at large” that is of the greatest interest to me and likely to you.
We’ve all seen them. In the context of the patent US 2018/0046717 A1 related entities are essentially this:
[image related-entities-1.png]
In this image the Presidents are the related entities but are not directly requested in the query.
The patent essentially discussed how these related entities are selected and how they are displayed. To further the example, if we click “Donald Trump” were taken to a query for his name that appears as:
[image related-entities-2.png]
Which differs from a direct query for his name without previously searching for Presidents:
[image related-entities-3.png]
We can see the breadcrumb navigation at the top of the results which they started in February of 2018 but more, we see the context carrying forward. When we searched for Presidents, a carousel of Presidents in chronological order was presented to us and when we click one the context is carried with it, something that does not occur when we search a President in isolation.
Ok – so what does this mean and what does it have to do with the patent? Let’s begin by digging into a few core areas of the patent itself. I’m going to save you the dryness of a full analysis of each section and simply highlight the key points. They are:
One of my favorite takeways is the reinforcement of what’s been discussed in previous entity patents and that’s the idea that there is an entity database. Essentially this is a separate database which is only tasked with understanding the various entities on the internet, what attributes they have and how they are interconnected.
For our purposes here we need to remember that an entity is not simply a person, place or thing (like Donald Trump) but also it’s characteristics. These characteristics are connected by relationships. If you read a patent the entities ae referred to as “nodes” and the relationships as “edges”. Some of the clearly prominent entities/relationships involved with Donald Trump are:
According to this and other patents (and logic), all of this is kept in a separate database outside of the general search index. Have I mentioned this is important? We’ll get back to this later.
In the patent we also touch on the important subject of determining relatedness. We discussed how relatedness applies to other areas in the previous article on optimizing for voice search. There are a few key ways that Google determines the relatedness of entities but one of the key mechanisms that comes up repeatedly is the co-occurrence of the entities in the same resources.
In our example above this would mean that the various Presidents would appear on the same page often thus indicating to Google that they are related. Alternatively one can assume that for our initial query “presidents of the united states” that each entity that appears in the carousel would be there regardless of whether they occurred on the same page together if they themselves appeared on a number of pages with “president of the united states” thus confirming that while Jimmy Carter may not appear on the same page as Donald Trump enough to connect those entities, he would appear enough times with his position to warrant the connection to the initial and thus subsequent queries.
This is also an incredibly important idea for general SEO outside of the patent we’re discussing.
An area of the patent that applies a little less to general SEO but is still worth discussing here is that Google needs a mechanism for determining which entities/relationships are most important. That is to say, Donald Trump is President – yes but he’s also a business person and would be connected to that entity by the relationship “Has/had job”. Why was this not selected as a primary characteristic?
One of the key mechanisms they use to determine which entity and relationship is the most important is the freshness (how recent are the co-relations we discussed above) as well as the clickthrough rate on related queries combined with what users type in after a query.
Basically, if people typed in “presidents of the united states” more often than “business person” the importance of that relationship would be increased.
Applying to all of these is the idea that authoritative sites, especially those related to a specific subject matter, are given a higher priority in determining the relationships between entities. So, a Wikipedia page on Donald Trump that discusses his role as President would be considered authoritative and strengthen that relationship more than this article here on Search Engine Land though here in this article a relationship between SEO and entities would receive the same treatment.
Think of it as PageRank for entities and similarly, there’s no green bar to tell you in you’re on the right track.
If you’re interested in reading the full patent you can do so here. But for now we’re going to answer the question …
So that’s the core of the patent in question and as noted above, there’s a lot that applies to general SEO and not just displaying related options within search results. Let’s look at how these areas impact general organic search …
If you couldn’t tell, having the idea of an entity database separated from the general search system reinforced in yet another Google patent strikes a chord with me and it should with you as well. You can think of it loosely like the database that maps all the links across the web to pass PageRank – only more powerful.
Instead of simply keeping a record of all the links and anchors around the web it’s taking it one step further and includes understanding the relationships between entities. This means that if you operate a hotel in New York and that hotel name is frequently referenced on pages with the entity “hotel” the relationship between the brand and the word “hotel” will be strengthened. Further, if it also exists on pages with “New York” that entity relationship too will be reinforced whether there is an active link or not.
Add to this, the mention being location on a site related to either New York and hotels and the relevance score is higher still.
Interestingly, being included on a page with other brands that are already strongly related to New York hotels would aid your efforts as well, essentially piggy-backing off the relatedness of their brand and passing it off to yours. And unlike PageRank which reduces based on the number of links, I have read nothing about diminishing relatedness of entities. This isn’t to say it doesn’t exist in some form in that a page about multiple entities could reduce the relative understanding of any one or more of them and that needs to be considered.
For example, having said hotel on a page with competing brands would (by this logic) assist in boosting the strength of the relationship to “hotels” however if the page is also about dining and activities in New York that may loosen the relationship. There is no information I know of at this time to suggest whether entity association is an on-and-off, relative-or-not scenario or whether the more entities referenced the less any one is valued. This would make sense and if that is the case then pages with a focus would logically reinforce a specific entity association more than a general page.
We do know that included in the patents in the notion that proximity to an entity is a signal, so the closer two terms appear on a page the stronger the relationship association is.
As with PageRank, authority matters. Unlike PageRank, the link doesn’t and if there’s a link – whether I’s nofollowed or not would be irrelevant. Now to be clear, I am referring to entity relationship building and not PageRank. PageRank and links are still powerful signals, but they are not what we’re talking about here. Basically, I’m not telling you to ignore link building or that nofollow links are as powerful as followed links (boy would THAT be horrible advice) but for what we’re covering here … nofollow would not play a role meaning Wikipedia and it’s nofollow links would pass a powerful signal. Some might even argue that all those new sites that have started nofollowing their links still have a high value provided the content and structure is presented in a way that the entity associations can be easily pulled out.
So, what we’re learning and having reinforced in this patent are some ideas about how to strengthen the association of one’s site/brand with specific terms/entities. The idea that we can push our rankings forward through entity associations and not just links is incredibly powerful and versatile. Links have tried to serve this function and have done a great job but there are a LOT of advantages for Google to move towards the entity model for weighting as well as a variety of their other internal needs (think about queries where the content is, by necessity, new and thus poorly linked).
Once more, I am not suggesting abandoning your link building. Not at all. This is meant to be in addition to that or even better, to prod you to focus your link building efforts on ways that can accomplish both tasks at once. If nothing else, you’ll be forcing yourself to pursue links on pages with a strong relevance to the attributes you want to be associated with so even if this patent and the others discussing entity valuation similarly are rubbish you’ll still be doing the right thing.