Understanding Entities with Dixon Jones

In This Episode

• [02:03] – Dixon Jones gives an overview of entity SEO and the importance of entities. He also explains how we can get a knowledge panel.
• [10:46] – How can InLinks help you implement entity SEO and schema markups?
• [15:24] – How would a one-person website owner set up its entities to grow their brand?
• [19:19] – Dixon gives an idea to start a knowledge panel.
• [23:24] – Stephan and Dixon discuss getting database entries in Wikidata and Wikipedia.
• [25:57] – Stephan shares about Kalicube tools, while Dixon shares how the owner of Kalicube helped him.
• [30:39] – How does SEO change as large language models become bigger nowadays?
• [36:56] – Dixon discusses his social media tool.
• [41:56] – Dixon advises on successful social media marketing.

Jump to Links and Resources

It’s great to have you back on the show, Dixon.

Hi there, Stephan. Thanks for having me back. 

First, let’s define for our listeners what an entity is and why they should care. 

Different people talk about entities and define them differently. An entity is a concept you can describe and put into a database. Now, a database is something like Wikipedia articles. I equate an entity as something that would have appeared in the Encyclopedia Britannica if it was still in paper form.

Walking is an entity. A big brand like Dyson is an entity. A notable person might be an entity. Part of it is, “Can you make yourself or your brand an entity?” The more generic part of entities and why it’s important for SEO is how you associate your brand with the entities or concepts like hiking, cars, restaurants, and whatever’s important. How do you associate your brand so closely with those entities that search engines and humans automatically associate your brand with the things you are selling?

How does one get a knowledge panel?

If data sources are done correctly, they can communicate very good ideas to a search engine like Google.

Getting your brand into the knowledge panel is about getting it to be an entity in its own right. How you do that, the perfect way is to get yourself a nice Wikipedia entry. Just go on, do it, and you can get deleted. That’s a quick way. If you can get a Wikipedia page, that’s great. It’s very difficult to do unless you have some superpowers in Wikipedia. 

More common is to look at a series of places that we think Google uses to help educate its knowledge graph. Things like Crunchbase and IMDb, if you are a personality, an actor or a podcaster. Companies House in the UK, for example, where companies are registered. New York Stock Exchange listings. 

There’s a whole load of other ones like Wikidata and other places with advanced directories. They are places of authority where you would register your business or be listed in a particular industry sector.

If data sources are done in the right way, those can communicate some very good ideas to a search engine like Google and say, “Hey, this is a whole list of licensed dental hygiene people.” And a search engine might take that list and say, “Okay, I can use that list. I can use that to test against other data sources that I’ve got to see whether this is (a) true and (b) of interest to put into a knowledge graph.” Building up a consistent picture of your brand across the Internet is vital. 

One of the big parts of that is using some schema to identify yourself on your webpage. You’ve got a place that is your authority for your brand. Even if you had a Wikipedia page, you might mention that in your schema to say, “It’s a Wikipedia page about my brand here.” Still, you’d put it on your homepage and hope that the Wikipedia page links back to your homepage at some point. That would give corroboration from two different data sources that the authority for your brand is on your homepage. You put some organizational schema in there. 

As important for me and from the SEO perspective, you can also use schema to say what the content is on the blog page or article that any user is reading. You can use the same schema to say this page is about hiking, and in case you don’t know what that is, Google provides the Wikipedia page for hiking. It’s also connected to this organization, Patagonia. You can sit there and help a search engine understand the fingerprint topics of a given webpage, and they know that you are an expert in that particular subject as well.

The trick is to be consistent and not spread yourself too thin.

The trick is to be consistent and not spread yourself too thin. If you’re a large organization like Amazon, you want to be known for your main assets like their cloud product, the fact that they deliver, or they’ve got online shopping. They don’t necessarily want to try with all their heart to be number one for used cars or number one for some of the products that they sell. That’s not their problem. That’s the problem of the people that own the shops or factories that make those products. 

Yes, you may still find Amazon in the results for buying second-hand cars, but ultimately what Amazon is good for is their larger topics, and then other organizations get more specific. You want to get very specific, whatever you want, and try not to do everything and be all things to everyone unless you are Amazon. They’re the ones that can dominate the world, together with Wikipedia.

The same as attribute. That’s not something that we spoke about in the previous episode. I know that that’s something that you want to, at a minimum, utilize with the schema corporation or organization markup so that you specify not just your social media URLs as attribute but also your Wikipedia article if you have one, your Crunchbase or AngelList.

You’ve got to make sure they reciprocate, so you don’t try to link your organizational page to a Crunchbase and then go back to a different page or cite a different page as the homepage.

That’s a problem that we’ve had at InLinks. We went from inlinks.net to inlinks.com. It’s not too bad because one’s a  301 redirect to the other, but it’s still not ideal because it’s not just the whole website. It’s usually a specific page you want to tell a search engine. A page that says this is the definition of your organization. 

I do, of course, use the same as for that, but there are a lot of tools that will help you do that whether you’ve got a Yoast plug-in, a Rank Math plug-in, or a bunch of other things. They will all help you set up that information about your organization or website.

What most of them don’t do is look at the actual content itself. When they do at the page level, they say, “Okay, is there some rich snippet to be had on FAQs here?” or “Is there some rich snippet to be had on the images or a video on this page?” 

What the vast majority of those automated tools are not doing is saying, “What is the actual content on this page, and how can we communicate that to a search engine? How do we turn this narrative of text into some very specific concepts?” There is a very valuable way of doing that using the same tag.

That’s where you do say this content is about this thing, which is the same as we use Wikipedia because we know Google understands Wikipedia well. But it means we want to prioritize the pillar pages and give them some context for a search engine over and above the organizational schema and the sitewide schema. You can do this very much at a brand level. 

You want to get very specific and try not to do everything and be all things to everyone unless you are Amazon.

Now, that doesn’t result in any extra rich snippets. Still, it helps a search engine understand more fundamentally the context on a page, which may or may not help your rankings directly, but it should and does seem to disambiguate problems. You are going to find yourself getting much more accurate traffic in there. 

Also, you get into Google Discover more quickly and tools like that where Google understands what this page is about. On the other hand, it understands what the user is interested in. Things like Google Discover will match those people up to your content, even though they weren’t necessarily looking for it that day. 

Let’s talk about how to implement all this. You’re going to use InLinks to accomplish this. 

You don’t have to. It’s automatically done in InLinks. In InLinks, we’ll show you all the topics your website has discussed. For example, you might have mentioned hiking thirty-two times on your website. If you don’t give some indication to a search engine as to which page is the most important page about hiking out of those 32 possibilities, then there’s a possibility that when a user is typing in ‘hiking,’ you’ve cannibalized your content and you’ve given no guide as to where the authority is for hiking as a concept within your website. 

By saying this as a human in the loop, you as a website owner or the webmaster, this page is the important one about hiking, which Google may have figured out anyway and hopefully did if you didn’t do a bad job. 

This is the page about, let’s say, hiking in the context of mountains. You’ve got two concepts there. One is hiking, and it’s hiking across fields. There’s another, hiking across mountains or up Ben Nevis, the highest mountain in Scotland. Ben Nevis is a concept, and so is hiking. 

You get the exact detail from schema.org. We say that schema.org can be about a thing, an event, a person, or a place. For this, we use things a lot called internal links. That’s the label. You can have it in any language. 

But essentially, you’re saying in the tags that this is about a thing called whatever it may be. There’s a label for it. If you don’t know what that is, “Google, here is the Wikipedia URL to define that.”

You could put any URL in there. This might be about a person called Stephan, and here’s the IMDB URL for Stephan Spencer. That’s entirely possible. The reason that we use Wikipedia is twofold. The first is it’s open-sourced and continually updated. If the underlying principle changes, the article will change eventually as well. The definition of what that thing is will get updated by somebody.

If a famous president dies, that will get reflected as long as you reflect a URL that gets updated, like Wikipedia. It’s open source, and we know that Google used it as a data source to this day and probably will continue to do so.

That’s a really good reason for using it. The other reason for using Wikipedia is that people misuse any shiny tool that SEOs are ever given. This is a good example. You’ve got to have a limit to what makes up an entity. You can’t just say everything. The world is an entity. 

You’ve got to communicate concepts that the knowledge graph will understand.

You might say property is an entity, finance is a property, and 10% is a concept. But you can’t say property finance at 10% as an entity. This is putting up more words and hoping some magic will work. As a human being, as a marketing person, to communicate properly with the knowledge graph, you’ve got to communicate concepts that are going to be understood by the knowledge graph. 

The best way to do that is to combine entities that are going to be actual entities rather than trying to make up your own, then find some mythical webpage out there and hope that Google is going to decide that this mythical webpage has the authority to be a reason to set up a new database entry for a new concept that’s never been in its knowledge graph whatsoever. Does that make sense? 

Let’s say an online coach, someone who is coaching about setting up an online store and is a one-person band. They have a small website with 20 pages. How would they do this without using InLinks?

If they are a one-person band, they would probably use their homepage to try and set themselves up as an entity or their organization as an entity on their homepage. But then, they would likely have a page for coaching, best practices for an online store, and many other articles. 

Let’s go for the coaching page because this is the one that is probably at the heart of their business. They say, “Okay, we need a tag to say that this content is the most important page on my website for the concept of coaching.”

They say this is about a thing called coaching, and in case you don’t know what that is, because let’s face it, Google could sit there and say coaching is about something to do with buses. It can make a lot of different definitions. There are at least two. 

Or baseball.

Or baseball, exactly. We’ll come onto that one. It’s about the concept of coaching. It’s actually about teaching. In the context of online stores, it’s about this, which is the same as this Wikipedia URL for teaching or coaching. Choose one and stick with it. 

Then it also mentions the context of online stores. It may mention Shopify and WooCommerce. You could also use the mentions tag to give it some secondary context. Then you’ve got four lines of code saying it’s about teaching and online stores, and it mentions Shopify and WooCommerce.

With those data points for a knowledge graph, a knowledge graph can see that there are only a fraction of pages in the world relevant to those four concepts with that priority. It rapidly helps a large data model to shrink down using numbers because those translate to numbers in their knowledge graph to find pages that have a very close imprint to the concept that the user is trying to look into, so when somebody’s trying to look at what’s better, WooCommerce or Shopify in ecommerce, then Google can say, “Okay, you need some coaching.” You can do whatever you like. You can think about it, come up with the concepts, and then try to find some data points that match it, and it sits there and sees that this page is well and truly about those concepts. It’s a close match. 

Do not try to go too broad in your content

That only works if you don’t get too broad in what you’re trying to do on the page. If you sit there and say, “I’m going to make this a review of all the different ecommerce platforms out there, and there will be 40-50 of them,” then that’s giving another message to the user and the search engines. 

You might then only get content back when it’s about ecommerce reviews, not learning about ecommerce. Being as specific as possible is the key, and not trying to go too broad in your content. It’s like keyword stuffing if you do it the other way. It doesn’t do much good at all.

Not because I did it but because I’ve been around. Let’s use you as an example if a person wants to get a knowledge panel. 

Mine’s rubbish, but I am there. If you type in Dixon Jones SEO, you will find it there. 

I put in Dixon Jones, just the SEO again. 

Yeah, and you’ve got a bunch of architects, no doubt. 

On the right-hand side, I saw results about Dixon Jones architectural firm and then Dixon Jones internet marketer. 

I’ve had this problem disambiguating for 25 years now, and the problem started when I bought dixonjones.com and Dixon Jones, the architects, bought dixonjones.co.uk. Unfortunately, the architects have been around for 40 years and designed the Royal Opera House in London. They’ve done some pretty impressive buildings, have a good Wikipedia page, and have some fantastic stuff. They got better when they left the business, so that’s good. I have to wait for Google to lose confidence in this company that doesn’t exist anymore.

An entity is a concept you can describe and put into a database. You want search engines and humans to automatically associate your brand with what you are selling. Share on X

It wasn’t helped by the fact that in the early days, I wanted to try and disambiguate myself. I had a page on dixonjones.com/architects, which, I could argue, was trying to disambiguate. What I was trying to do was get the traffic and then disambiguate. That page said, “I am not Dixon Jones, the architect. If you want to talk to them, they’re over here, etc.”

While doing that, I associated myself with Dixon Jones, the architect, but not with the authority of the actual homepage. Of course, Google got it right most of the time, and when you type Dixon Jones, the architects always came up top before the knowledge panel ever existed.

Even on relatively simple search engines. If you go to Majestic, which has a search engine and type in Dixon Jones, it will put the Wikipedia company up first or used to. It may not be now because they’ve come offline.

Trying to get my panel in there, what I did was I became a product leader on Crunchbase. I wrote a book. Writing a book is a good way to give yourself a knowledge panel. I think it’s not so much getting the book on Amazon that helped. It ensured it had a proper ISBN or you got an American equivalent. Ultimately, it was getting into Goodreads, not getting into Amazon, that gave me the panel. 

Then it’s a case of, “Can Google match everything up?” Then say this book over here is the same as this webpage over here, that the homepage is the same as this Crunchbase profile. Now I also have an IMBD profile for a podcast I used to do and I used to write, which is probably not a good thing to associate with me because it’s so far off what I want to be known for now, but the podcast was SEO-related, so that was okay. 

These are the things I’ve got together. I would say that I’m not particularly noteworthy. But I’m enough to get a knowledge panel, and if I can do it, anybody can do it because I have no importance in the world except trying to influence search engines a little bit. 

Writing a book is a good way to give yourself a knowledge panel.

I would doubt anybody would see my knowledge panel unless they were also related to the SEO or Internet marketing industries. People on this podcast who don’t know about it are not seen daily as interested in the topic and may not see me in the knowledge panel because Google is trying to match the user query with the user intent. The user query only talks about Dixon Jones in the concept of SEO if the user is interested in SEO or internet marketing.

You’re saying that Goodreads was probably a pivotal accomplishment for you getting in there with your book. Not necessarily getting into Google Books because I see that as a source in many knowledge panels for authors if they’re not notable enough to be in Wikipedia and don’t have a Wikipedia page, but they have a listing or their book is listed in Google Books.

I think one dictated the other. I think Google picked up Goodreads. Then it went to Google Books. I don’t recall going and registering on Google Books. Google seems to pick this up from third-party data sources, including Amazon. I felt that Goodreads and the knowledge panel came along very closely. 

Okay, I got it. Did you get a database entry in Wikidata?

I did. That comes as an automatic thing or aspect of Goodreads. I had one in there anyway because I tried to disambiguate Dixon Jones, the architects from Dixon Jones, the internet marketer, and that worked for a while. Then eventually, Wikidata said we needed some more notable articles for Dixon Jones, the SEO, and they were not happy with the ones that were put forward. Goodreads then generated it automatically. That became great because I could refer to an auto-generated Wikidata article. 

However, because of the first one, I think the whole lot’s been deleted now, and I don’t think it’s in there, but I might be wrong. I haven’t looked recently, but I feel that I’ve got a page in there the last time I looked. I was talking with a human editor there who said to come back with some better-cited sources, and we’ll put you back in. 

Building a consistent organizational code and schema can help ensure your brand is properly represented. Share on X

I don’t know of any requirements for Wikidata or Wikipedia, and they’re not necessarily as stringent as Wikipedia, but—

They still exist, and I don’t want to push the envelope because I’m not that notable at the end of it. I would rather concentrate now on trying to be, and I suppose, more notable. No, the narrower focus is the SEO and the SEO that manages SaaS. I’d be notable for the things I am known for and try not to mess that up. I think that I’ll get there eventually. 

As you see, I am there. It knows where I went to university. It knows the date of birth. It knows many other things about me that I have seen in the knowledge graph. But I think there’s also a confidence level that Google goes through as to how confident it is that it’s the right answer for the right person at the right time. Frankly, I haven’t got much confidence in myself whatsoever. 

There’s a tool on Kalicube. 

I’ve used that a lot. I should have said that. I think Jason does a brilliant job of citing the data sources that he or his systems have seen in knowledge panels. The argument goes that if Google has used a particular data source and put it into a knowledge panel, then why not go and make sure that your records on those data sources are kept up to date? It’s an effective tool to do that. That helped me get into the knowledge panel using Kalicube’s tool.

There’s a confidence level that Google goes through regarding how confident it is that it’s the right answer for the right person at the right time.

Kalicube has several tools. If you go to kalicube.pro, you’ll see it listed on the top nav, such as the knowledge graph trusted sources list. Then the knowledge graph explorer tool, which gives you if you put in a keyword, I don’t know how it does this, but it somehow gets these IDs back of knowledge graph entries, even if there’s no knowledge panel showing up for that search query.

If you have a low confidence level or Google has a low confidence level of that entity and doesn’t show it as a knowledge panel, it’s still in the knowledge graph. You can see it there in Kalicube’s tool. 

I found the tool’s paid version useful, but I needed Jason’s help. Jason, the owner of Kalicube, helped me quite a bit. It tries to read the entries and tell you where there are disconnects in joining the dots, and you can then go and improve your systems. It was amazing how much you could pick up citations of yourself, or if it’s your brand, then your brand and work with those.

However, what was also interesting is that the search engines pulled out a lot of wrong stuff, and being able to tell or build code that disambiguated was useful to make sure that you were not associated with, in my case, an architect, is helpful as well. 

Although I think all it ultimately did was generate your organizational code for you. Basically, or in my case, your schema is to say, “Hey, this is a person called Dixon Jones, who is the same as these profiles worldwide.” It gave you the code for that, but it also allowed you to just jump over to these places and get registered on some things that you may not have seen. 

Or you were seen, but you had an old URL. I used to be the founder of an agency called Receptional. The agency still exists. I think they’ve eradicated all mention of me, but I might still be on some systems somewhere, and they don’t particularly want to be associated. It’s not that they don’t want to be associated with me. I think they are still happy to call me the founder, but they’ve got their own culture and moved on, and that’s fine with me. I don’t want to be linked through to say Dixon Jones is CEO of Reception because I’m not.

When you were referring to Jason Barnard, a past guest on this show—if you’re interested in learning more about the knowledge graph and getting knowledge panels, there is much more in that episode—the kalicube.pro tools are not just the knowledge graph explorer and trusted sources list. He’s got a brand tool, I think. The knowledge graph sensor. Some brand surf explorer or something. Anyway, there are some great tools.

This whole idea of entities is really interesting that three or four people have jumped out and tried to find a way forward with it. One of them is Andrea from WordLift. He came up with a tool that looks at content. Jason looked at this idea of your brand being the entity and how to make your brand stand out properly in all things digital. I approached it from this idea of the more generic concepts and associating yourself with those concepts.

We think they all dovetail with each other well. Still, they’re slightly different ways out of the starting gate, I suppose, for the idea of an entity-based approach to search engine optimization anyway but also to brand building, and they’ve worked very well. 

I’ve also had Jason on my podcast, and I’ve been on his. We’ve talked quite a bit about these kinds of things. And interestingly, we had to wake up one day and say Google’s changed here, but we don’t know how. No one told us how, but we can’t stop. It’s going to happen all over again with LLMs. 

Large language models, yes. 

Absolutely. I think that’s going to be the next challenge. Not all the work they’ve been doing with entities has disappeared, but I think that large language models and natural language processing overlap well. I think the work we’ve been doing at InLinks can map very nicely onto how learning large language models will work and hopefully will be an extension. 

How does SEO change now that large language models such as ChatGPT and Google Bard are so much part of the world?

This is where I have to get philosophical, Stephan, because I think that the interesting thing about these models is that they’re based on reading millions and millions and or billions and billions of docents and getting a consensus of ideas. It’s going through billions of documents, tying them into tied concepts. Some knowledge panels connect dots, so they sometimes have trouble citing their sources. 

The way that they’re learning is much more like looking at information in the same way that our brains look at information. Our brains are made up of billions and billions of synapses connected, and the way Jeanie, my daughter, now talks about SEO as well because I’m that old in life. Her degree was in language, culture and communications. 

We as individuals will grow up looking at life differently.

She learned schema and talks about the schema in terms of a guy who was in the 19th century and talked about schema as an extension of the way the Greeks looked at schema, which is saying the way that humans are making up their understanding of the world each person individually is waking up one day and finding new experiences. Every new experience will potentially alter their view of the world.

Her example is you wake up one day and are a kid. You go into a field and see something with four legs, hair, and a big tongue. You’re told it’s a cow and say, “Okay, now I understand. Now I know everything about a cow. I’ve seen one in real life.”

The next day you woke up, and then you went and pointed to a horse and said, “Hey, Daddy. There’s a cow.” Of course, it’s not, and you have to correct your schema, and you have to work out the difference between a horse and a cow, and you have to address that. That is constantly happening in the human brain. 

It means that we as individuals will grow up looking at life a little bit differently. My example in my talk is, “Can carrots help you see in the dark?” If you ask the average Brit, they’ll say yes. If you ask the average American, they’ll say, “What?” 

The truth is during World War II, Winston Churchill’s war machine was trying to do two things. First, they were trying to get the kids to eat carrots because that’s all their food. Secondly, they were trying to explain why Spitfires could suddenly find the Luftwaffe without giving away the fact they had radar because radar was a secret weapon for them.

They were using this to say we give the pilots loads of carrots so they can see in the dark. That propagated on through generations of Brits who tell their kids that carrots can help you see in the dark. Any other country didn’t get that schema, and they didn’t get that concept of the world, so they have a different view. Everybody else’s view is correct, but until I was quite old, I genuinely thought that vitamin A or D could help you see better in the dark. It turns out that whether it’s true or not, the reason that I came up with it was completely false. Correlation may not be causation. 

I think that for LLMs, this is the same methodology. These systems will come out with outright lies or things we think are lies. Still, they are just a formation of their schema based on what they’ve seen, which reflects what we’ve all written as human beings over the last few hundred years because these things have gone and read all the books of the last few hundred years. 

If we as a society say things wrong, then that will make the ChatGPTs of this world get things wrong if we start to do it at scale. 

Associate yourself with authoritative entities. Google often gets it right, but it’s crucial to ensure your online presence is aligned with reputable industry-related entities. Share on X

If we then go back to the 1960s to how advertising and marketing worked, it wasn’t how to get into the head of the user. It was all about TV coming online. How do you get into the psyche of all of the audience out there? I think that’s what’s happening now. If we start to think about this and optimize for ChatGPT properly, I see how we get it so that our brand or the things we stand for are talked about when somebody asks a related question to ChatGPT. 

At a superficial level, one of the best vacuum cleaners, and you would hope that Dyson, Hoover, and Shark are on the list of what comes back. That’s very specific. What’s even more impressive is how I clean up the dust on my carpet. Then if a Dyson, Shark, or Hoover gets listed in that answer, I would say that some SEO somewhere has done a good job.

To do that, you must return to ensure your brand is closely associated with a narrow,  well-defined idea, carpets, hoovering, or vacuum cleaning. Hoovering has done it, so hoovering is already a verb. Try to make yourself a verb in this process. 

Not so much here in the US. 

If we as a society say things wrong, then that will make the ChatGPTs of this world get things wrong if we start to do it at scale.

Okay, in the UK, we all grew up saying, “Get the Hoover out,” and now we might say, “Get the Dyson out.” But certainly, growing up, Hoover was a thing, a concept. I think we need to work on that. Optimizing for ChatGPT and those systems is all about the same thing as the madmen of the 60s trying to convince a whole generation that a particular brand was the right thing to associate with butter or toothpaste. 

If you don’t do entity SEO, link building or technical SEO on your website, you will make it much harder for the large language models to know that you exist and are part of the global conversation and your needs for the topic.

Yeah. It’s like digital oil, isn’t it? You’re oiling the system. 

Cool. Now let’s discuss your new social media tool because we haven’t discussed this yet.

That’s just come out. We’re taking some of a pump with this, but I think it’s the right idea. The idea of coming out with a new Buffer, a new Hubspot, Sprout Social, or whatever your favorite tweeting technology is, initially for me, seemed to be a dangerous thing for us to get into. It’s a big world with a lot of players in it. 

We realized that none of these tools were creating a content plan for tweeting and Facebook posting and a sensibly curated LinkedIn posting. By that, most of the tools you use just write your post. That’s it. You just wake up and type into it, and it’ll post on those three things next Tuesday for you. Cool. 

Knowing what to write was a lot harder to get right. Maybe not for an individual, but for a brand, it’s very difficult because they may have many stakeholders, people, customers, values, and expertise that are possibly siloed all around the organization.

Knowing what to tweet about, especially if you want to tweet every day or LinkedIn post every day and you want to come across as a brand that is a knowledge leader in your field but not a knowledge leader outside your field, then you can’t just make up some words and say give me a whole bunch of things and go to open AI and create a bunch of tweets. You need this step in between that gives you a list of things to tweet about or post about that are coherent with your existing strategies.

We said I think we’ve got something here because we already know all the topics your website is discussing. What we can do is take all these topics, then we can generate more ideas. We can use Google Suggest to generate more ideas around there. What are people looking for around the concept of vacuum cleaning a carpet? 

You can put a vacuum cleaner into Google, and Google Suggest will come out with a whole different thing about the price of a vacuum cleaner, things to do with the vacuum cleaner, or how to repair your vacuum cleaner. 

But it may also come out with things that are not appropriate to your content. The example I use is Mustang. If you put in a Mustang, there’s how to buy a Mustang and how to service your Mustang, but there’s also how to feed a Mustang and what kind of grazing you use on a Mustang. There is also some stuff about Mustangs involving airplanes. 

Mustang has three different concepts, and when you go to Google Suggest, some of those will be, unless Google’s got another context like knowing your history and your predisposition to horses versus cars, it’ll come up with suggestions. You need to put the suggestions through the loop then and see whether the suggestions coming out by the system are closely related to the content or on the website. 

We need to be mindful of the accuracy of our generated information to prevent large language models from perpetuating false or misleading claims. Share on X

The result of doing this runs for a fair amount of time. But on your site, we can throw it all back in, so we can take your website and say right from it, here are all the important primary concepts. We cluster those up. These are the interesting questions that people are typing in there and want answers to that are semantically close to your existing content. We’ll also pull out Q&As you’ve already put on your website and say, “Right, here’s a tweet and an answer you’ve already written, so these are there.” 

For an average site, let’s say it has a hundred pages, and we will have thousands of ideas. We’ll have them clustered by concept as well. You’ve got maybe half a dozen concepts, and within each, you’ve got ten or twenty different things you could tweet about. It comes up to thousands. 

You can then say, “As a human being, all that work’s done for me. All I had to do was put in a website.” Click the ones that I think are important. Some of them will be very similar to each other. You can delete the ones you don’t want just with a click, which builds a content plan for the next 52 weeks of ideas. 

Having all those ideas put out in front of you built out of your existing web presence removes a huge amount of bias in a personal social media manager’s decision-making process. A social media manager that can’t suddenly read all the websites and make sure that they grab all the ideas and throw them back in, which would be thousands of hours of work, is coming back with a plan that is now going back.

When it comes to a brand, you need consistency and completeness in your social media marketing.

You’ve got logic for those ideas, and you can then get buy-in for that plan based on some science rather than, “Oh, you know what? We’ve got the Pride Festival over here.” We should probably do something about pride or just come out with an idea that seems like it’s going to be catchy. You can do that too, but when it comes to a brand, you need consistency and completeness in your social media marketing. 

So coming out with that, you create the posts based on that. You agree with the plan. Then we allow you to write your own, or you can press the button and allow ChatGPT to do the first draft, at least for you, but that’s not the prime bit of the tool. The prime bit was getting that list in the first place and making a much more scientific judgment there. 

Then the tool will allow you to say, “I’ve posted it. We now want a call to action. What’s the webpage on our website that’s most appropriate for this?” You can choose it from the dropdown. If you want to add an image, you can either upload your own, or it’ll link into Pixabay and use open-source photos, if that’s what you want to do, to give it a bit more pizazz. Then you can schedule it.

It does the same things that other social media schedulers do. It’s that prebit of what is the hell I do, not just tomorrow but for the next six months in my social media plan, and how do I connect these dots? By doing it this way, we are tying it back to the SEO that’s been done on the site, the content that’s been done on the site, and the brand itself, so we feel that we are going to develop somebody’s social media strategy in consort with their SEO strategy and in consort with the concept of entities, things not strings. 

That sounds cool, and that’s a very different approach from, let’s say, going to also ask.com and having it aggregate a bunch of questions based on people also asking boxes in.

Yeah, it is different. Mark Williamsons-Cook runs alsoasked.com, and I’ve got a lot of respect for him. We share a little backroom where we cry about the issues with SaaS when the credit card stops working. We get on quite well, but you’re right. He does different things. It’s a different approach. 

And they both have a great place in your strategy. Congratulations on the new tool. 

Thank you very much. I hope people try it out. I would only say you can use it on the free plan at InLinks, but you need to have at least 20 pages in the system before we start thinking. On a free program, you only get 20 pages. You’ve got to put in 20 pages of your website before we’ve got enough ammunition to go and start finding the ideas for the plan, but it can be done. 

Awesome. Are there any parting words of wisdom we haven’t already discussed with our listeners? 

My word of wisdom for the day would be that we’re all wise. It’s just a question of what. We can’t all be wise at everything; whoever we admire in life will only be in context. 

All right. Awesome. Thank you so much for coming back on the show, sharing what you’re up to, and what can help move the needle for our listeners. 

Thanks very much. Thanks so much, Stephan. 

Your website is inlinks.com. 

inlinks.com. Now, it may have been inlinks.net when we last spoke. We had a very nice customer who pretty much gave us the domain. I’ve given him a free account for as long as he wants. 

That’s awesome. Congratulations on your company’s success and forging out on your own after Majestic and creating a great SaaS tool.

Thanks a lot. 

Thank you, and thank you, listener. Get out there and do some entity SEO. We’ll catch you in the next episode. I’m your host, Stephan Spencer, signing off.

Important Links

 

Apps and Tools

 

Book

 

Businesses/Organizations

 

People

 

Previous Marketing Speak Episodes

 

YouTube Videos

 

Your Checklist of Actions to Take

Discover the power of entities for your brand. Store relevant concepts in a database to enhance search engine and user association. Elevate your brand’s online presence by understanding and leveraging entity connections.

Establish a knowledge panel to develop my brand as its own entity. Creating a Wikipedia page for my brand is a powerful way to achieve this.

Harness authoritative sources like Crunchbase, IMDb, and industry directories. Registering in reputable directories boosts search engine visibility and provides valuable brand information.

Implement Schema Markup to identify my brand on my web pages and provide structured data about my content. This helps search engines understand and display my brand more accurately.

Concentrate on being known for my main assets and areas of expertise. Instead of trying to dominate every area, I specialize in my core offerings to establish a strong brand reputation.

Keep learning and stay informed about the latest industry developments, algorithm changes, and SEO best practices. Entity SEO is an ever-evolving field, and it’s vital that I keep up with it. 

Help my brand to stand out from unrelated entities. Incorrect associations can lead to confusion and dilute my online presence.

Create a content plan specifically tailored to my brand. Social media planning involves knowing what to write and ensuring it aligns with my brand’s expertise and strategy.

Analyze my website’s existing content and cluster important primary concepts. I can use social media tools to do this, such as InLinks. 

Visit Dixon Jones’ website to learn about digital marketing, link building, entity SEO, and more.

About Dixon Jones

Author of Entity SEO and CEO of the entity based SEO tool, InLinks. Previously CMO of Majestic and a long time SEO Speaker for over 20 years.