Scott Services: Weekly Wisdom with Jason Barnard: Knowledge Graph for Brands

Hi, I am Jason Barnard, and welcome to Weekly Wisdom. This week, I will discuss how to get your brand into Google’s knowledge graph. Here is a quick intro.

What is the knowledge graph?

A very simplistic explanation is that it is an encyclopedia for machines that they can understand and use independently; think Wikipedia for machines. The knowledge graph is absolutely fundamental to how Google functions today. It is the underlying foundation of Hummingbird; it is the key to entity-based search, and it is, therefore, essential to all your SEO efforts. If it isn’t already, the knowledge graph should be a big, big part of your SEO strategy. Start now working on getting your brand into the knowledge graph.

So how does it work? It is really simple. Entities and relationships. Here is a really nice, simple example. Mary knows Kate. Mary likes Pete. Pete is the brother of Kate. Pete was born on the 27th of March 1982.

Now, that was really easy for you to understand because that is how the human brain functions, and here is a really fun demonstration. I met a guy called Chester. He was giving me a trick about how to remember 15 random words in three minutes. The random words were: monkey, iron, rope, kite, house, paper, shoe, worm, envelope, pencil, river, rock, tree, cheese, dollar bill.

Now you will notice I could remember that, but I had to think about it. The reason I can remember all of these words is because he made a story. He said, “it is a fun story,” and that was what he was concentrating on. What I realized is it isn’t 15 random words; it is 15 random entities, and what he did was join them together with relationships. Here’s the trick:

The monkey is holding the iron. The iron is attached by a rope to the kite. The kite crashes into the house. The house is covered in paper. A shoe walks on that paper. There’s a worm in the shoe. The worm jumps out into the envelope. A pencil writes on the envelope, the pencil leaps into the river that splashes onto the rock. The rock leaps out to the river, hits the tree, the cheese falls down. It is a cheese tree. The cheese breaks open, and a dollar bill falls out.

What he has done is taken 15 entities and made relationships between them that makes it incredibly easy for the human brain to remember.

In that example, we had lots of nouns and verbs, and nouns and verbs are entities and relationships. Subject – verb – object = equals – entity relationship -entity.

And the latest Google update, that is all about Bert, B-E-R-T, has made this much easier for Google to see subject-verb-object and extract from those entity-relationship-entity. If you want to know more about Bert, look up what Dawn Anderson is saying. She is the go-to person for Bert.

We have the knowledge graph, which is an encyclopedia that expresses entities and relationships that Google uses to understand the world, and Google uses also uses it as its underlying foundation to its ranking system. Sooner or later, if your SEO strategy is to succeed, you have to get your brand in there. Here, I will look at how to make it sooner rather than later and get the jump on your competition.

Now, how do we add information to the knowledge graph?

Chester was asked the question, “Isn’t it the case that if I remember all of this useless information about the monkey and the iron and the house and the river and the cheese tree, that I will forget something more important?” And he said, “No, the important thing about memory is it expands, and it expands. The more entities we have in our mind, the more possibility there is that a new entity can be attached to a relationship with another entity that’s already in there, and that makes it much, much, much easier for us to remember and to understand.”

So, taking this a step further, when you learn about a new entity, if you have no existing entities that have relationships to it, it makes little or no sense. And it is the same thing for Google.

Further still, the more entities you can attach to this new entity with relationships, the more sense it makes to you, and the better you can remember it. The same goes for Google. So that is how you can get into the knowledge graph.

Find entities that are already in there and barnacle on them, use them as hooks for understanding for Google. Now, the question is, how do you find those entities that you can hook onto to push your information into the knowledge graph. With Wikipedia pages, that is the very obvious example, but not all entities in the knowledge graph have Wikipedia pages.

This tool from Kalicube.pro is a good way to check a presence in the knowledge graph. Unfortunately, Google’s API lags behind what they are actually using, and it is, therefore, a little bit incomplete, so if the tool doesn’t return anything, that doesn’t mean to say it is not a candidate.

You can also look for the knowledge panel on the Google results for that entity. Search for the entity — if the knowledge panel turns up, the probability is, it is in the knowledge graph, but be careful of featured snippets.

On that right-hand side, there are a lot of featured snippets that are disguised as knowledge panels, so watch out for featured snippets that are disguised as a knowledge panel. Sometimes on the right, you get a featured snippet, and it looks like a knowledge panel, but it isn’t. The sharing symbol is what indicates that it is a knowledge panel and not a featured snippet.

Now, a big hint for choosing the right entities. The stronger and closer the relationship is, the more leverage it gives. So choose entities that you have strong and close relationships with.

What do you need to do to get your brand in the knowledge graph?

You need to explain to Google the relationship between your brand and multiple entities the knowledge graph already knows. Step one, and this is really step one. The first thing to do is to do that on your own site. Write a piece of content that explains the relationship or include the relationship in an existing piece of content such as your homepage, the about us page, or your partner’s page.

Pay special attention to indicating clearly the type of relationship, the strength of the relationship, and the closeness of the relationship. To reiterate, the closer and the stronger, the better.

Make sure you add schema.org markup to communicate the relationship explicitly in a format Google can digest easily. Step two is corroboration on trusted sources. Wikipedia is obviously very helpful, but it is not necessary. But that appears to be a commonly held belief, but it is false. You don’t need Wikipedia. Wikipedia is simply very helpful today.

So add your entity to Wikidata, Crunchbase, any specialist data sources you can find, IMDB, Google Books, a registry of dentists, whatever’s appropriate for your industry.

Step three is to point out to Google the corroboration on your site. Add links to these references from the content where you indicated the relationship back in step one. If there are articles or major publications that confirm the relationship, Google is getting better and better at extracting entities and relationships from content.

If you think back to Bert that I talked about earlier on, that is incredibly powerful. A big powerful media site is a great source because it is trustworthy. All Google needs is to be pointed in the right direction. And once again, at this stage, add Schema markup to be explicit and make things simpler for Google.

Also, add links between the citations that corroborate wherever it is possible, and that creates a lovely web of entities and relationships with your site right at the center.

And that begs the question, how far can you push this?

The answer, infinity and beyond. Once you have that web for your brand, you can build that out. You can spread your web to pull other entities around your brand into the knowledge graph. Products, people, places, events — anything you want to get Google to show to its users.

The more you do this, the stronger your web of entities and relationship becomes, the wider do you can spread that web, and the more you can leverage it for your SEO efforts. Lastly, there was a huge update to the knowledge graph in August 2019. It appears to be a real game-changer. If you want to know more about that and you should, if you are doing SEO, read this article on Search Engine Journal.