Behind the Scenes with Dixon Jones: The Unseen World of Entity SEO

Mark A Preston: Welcome to the unscripted SEO interview podcast. I'm your host, Mark A Preston, and today we have Dixon Jones, the author of Entity SEO, the CEO of Inlinks, and the global brand ambassador of Majestic. Please welcome Dixon. Hi, Dixon.

​

Dixon Jones: I'll give the clap because there's no one here, so how are you?

​

Mark A Preston: Very well, thank you. Right, first of all, what's your SEO story summed up in all these years?

​

Dixon Jones: I used to write and run murder mystery evenings actually, and I used to have to send my actors out to wherever they were, and it was a pain in the neck. Every night, we would get home at 3:00 in the morning, and it was exhausting. When the internet came along, I thought, "Oh, maybe I'll get a website." Instead of getting a website from wherever everyone else got one from, I got one from the ISP where I could see the building. I didn't really understand it, and I thought, "You know what, I want to knock. I mean, it's Windows 3.11. I want to knock on the door and get my hard drive if I need the data back." That was the kind of mentality I had. The only thing that ISP had going for them was that every Thursday, WebTrends would run a log file analyzer.

 

So every Friday morning, I could wake up and see where my visitors had come from, what search engines had sent them, and this kind of stuff. I found people coming to my website from Ohio and other places, and I thought, "This is really weird because I can't get my actors to Ohio, so how do I change my model? How do I work it all out to make money out of that?" So I created a downloadable version of one of my murder mysteries and started selling it online. Then somebody tried to sell me MurderMystery.com for lots of money, and I thought, "I could get MurderMysteryGames.co.uk for £2 or £10 or whatever it was." Then I kind of figured out that I was selling murder mystery games, not murder mysteries, so that was a more targeted kind of idea. I started building search engine-optimized websites, and then I got a call, and I built them in FrontPage because my skills were not particularly good, and the FrontPage Users Group called me to a meeting. And my first SEO talk was at Microsoft's HQ, which is pretty cool. So that's how it all started, really.

​

Mark A Preston: When was that? When did you start speaking about SEO?

​

Dixon Jones: Well, I was fighting with FrontPage 97, so probably around there. I would say that FrontPage 98 was when I actually got it to work properly. I can't remember if I had FrontPage 95 or 97, but anyway, let's say 98 is when I really got into it. And then in 1999, I set up a limited company. So I founded Receptional, but I no longer have any part of it. I'm not speaking for Reception anymore. Founder or co-founder is the only word I can use to describe my relationship with Receptional. But they're still going today and much more successful than when I ran it. But that started in December 1999, so yeah, I was a full-time SEO by December 99.

​

Mark A Preston: So were you working with your own clients or did you work elsewhere?

​

Dixon Jones: One of my first clients was actually not SEO. It was for a major car company who wanted to know what changed on their competitor’s websites every week or month. I don't know, I think we did it every week. Every week we kind of crawled all their websites, all the competitors, the Fords and the Toyotas and stuff, and then looked at what had changed on those pages and then did a report of what changes, particularly price, but also images and stuff like that. So they were keeping up the competition. So that was pretty cool because it was good money. Bit of a slog. I mean, we used some tools. There was one called Morning Paper, and there were a couple of other tools that would check the differences between the code. But a lot of the sites at the time were flash, so all we could do, really, is print screenshots of those. We couldn't even print the page and then have a look and see if they changed. So that was one part of it.  The first major client, also in 1999, was actually my wife was working for Mohammed Al-Fayed, who was the owner of Harrods. But she was actually working in a block of apartments, the luxury apartments in Park Lane. And they spent quite a lot of money on building a website.

 

One day they asked me to have a look at it because I'd already optimized murder mystery games and stuff, and I remember I was supposed to be going for Christmas drinks on the Thames, I think it was, but I got stuck with the general manager of this place for an hour or so, talking through, explaining why Frames was not ideal for SEO and stuff like that. So after about an hour, he just looked at me and said, "Can you fix it?" I said, "Yeah, I think so." Basically, I converted it because it wasn't even frames back then. So that was my first real customer. Then, to the credit of the designers, the people that designed the website, which were quite a big agency, when they saw what I was doing to the website, they then called me in for other customers as well. So I got other customers off of them as well and that really kind of set the ball rolling. But I started with the SEO consultancy rather than agency. I didn't know what an agency was, frankly, I just knew that everybody on the internet really needed to understand the SEO stuff. Because on the murder mystery game stuff, if I went to Yahoo, as was the big one at the time, and typed in murder mystery games, there was a big banner ad for Adidas at the top. So it was completely obvious to me that firstly, Yahoo got their business model entirely wrong, and secondly, there was a real opportunity if I could just reengineer the algorithms. So that was it.

​

Mark A Preston: Since you started up to the present day, what would you say has been the major changes within the SEO industry?

​

Dixon Jones: Well, it all changes so much and not at all, doesn't it?

​

Mark A Preston: That's why I'm asking, because has it really changed, or what are the significant...

​

Dixon Jones: The principles of marketing have not really changed. And I think that in the early days, people were trying to get as many customers, as many visitors as they could, and then figure out how to convert them afterwards. And I think that was a bad approach for SEO. And I think that's the approach that got SEO into the quagmire that it got into in the early 90s, and pretty much until Google dragged us out of the mire and made us some degree of respectability, I think. But those that were cognizant of how SEO fits into a proper marketing plan, I think probably made fewer mistakes. We all made mistakes along the way, no doubt. But I think that the idea of getting the right customers instead of the idea of getting all the customers is the way to go, and always the way to go. But it took us a long time to figure that one out as an industry.

​

Mark A Preston: Some maybe haven't figured it out yet.

​

Dixon Jones: No, I'm sure the spamming will go on, and as long as there's a business model that pays per ad view, I'm sure that will carry on going forever. But I think now SEO is more about disambiguating what your content is about than it is about ranking for generic keyword here. It's about finding the point of difference more than it is about finding a sweet spot that all of a sudden the whole country wants to dive in on. And I think that has always been a principle of marketing. You've got to have your point of difference, your USP, your value proposition, whatever it may be.

​

Mark A Preston: So you mentioned you do a lot of speaking around the world, traveling. (Dixon Jones: Yes.) What sort of topics do you speak about?

​

Dixon Jones: Entities? Well, I talked for a long time about backlinks because being ambassador to Majestic and then previously to that, I was marketing director for Majestic. And basically when I joined Majestic, it was a couple of guys in the back room. So it was very early days when I joined Majestic. Talking about backlinks has always been a big part of things, internal linking now with Inlinks, and also schema, specifically entity-based schema and entities is really where I moved from. Around about 2014, when Google bought MetaWeb from Freebase, my brain was starting to think, "How come they've spent all these something like hundreds of millions on a database system that doesn't even produce any money yet? What happened there?"

 

And it took me a long, long time to think it all through that ended up in the book "Entity SEO" because I was just trying to figure out firstly what knowledge graphs were. But really, for SEOs, it's not so much about getting yourself into a knowledge graph. It's how does that help your business or your company or your client out of that? So the trick is not getting in a list. You could always submit yourself to a million directories. The trick is getting a customer to become more aware of your brand, or to buy your product, or to engage with your ecosystem, and trying to get somebody from somebody else's ecosystem, Facebook's or Google's or the TV ad, whatever it may be, into your ecosystem where you could communicate with them directly on a one-to-one basis, where you could help that person to hopefully make a decision that is in your favor about a product or a service. So I kind of moved from talking about just links and things to start to talk about entities anytime after around about 2014, I suppose.

​

Mark A Preston: So, regarding entity SEO, how would you explain that to somebody that doesn't know anything about SEO? What is it?

​

Dixon Jones: And the thing about describing a page of content in terms of entities is you suddenly cut down the amount of memory that is needed to store that information. So, if you imagine a search engine trying to read a page of text and understand it, in order to do that, it really needs to jump from idea to idea to idea to idea. So, if you turn it into a bunch of numbers, which are lookups in an encyclopedia, you can see how when you're trying to organize the world's information, a bunch of numbers together in a certain order is as good as a fingerprint for defining what's on a page. But also, if all those numbers represent concepts and ideas, then the numbers themselves tell a story. You could almost take those numbers and then translate them back in any language, and you'd have a very similar story. So, it became a very powerful and alluring idea for me to say, "Okay, if Google has decided we're going to buy Met web and turn everything into a database, why would they do that? How can I then influence that? What do I need to do?" I need to get my brand, my client's brand, or my brand into a place where it is constantly showing up around the topics that are important to me.

 

So with Majestic, I was very keen on making sure that we were authoritative around the concept of backlinks. And with Inlinks, again, internal linking Schema and entity SEO, as we now call it, these are very important concepts. But if I was Ford, then it was very important that the Mustang gets associated with cars instead of getting associated with horses in the wild. And that's where I get to the Disambiguation bit, is that really your brand has to spend a lot of time disambiguating itself. Even today, I've lost my Wikidata entry for Dixon Jones. There's an architects firm that is now bankrupt, by the way, that used to own Dixon Jones Co UK. And I used to own, well, I still do own, Dixonjones.com, and in the early days when you're trying to get all the information, all the traffic at dixonjones.com, I did /architects and had a big page on why it was disambiguating. It was explained that I wasn't an architect and that there was a famous firm, a firm of architects, but I was trying to rank for Dixon Jones architect for no logical reason except that Dixon Jones Co UK was outranking me on Google, which was very annoying to me. So, disambiguating is really important. But they've got a Wikipedia entry, even though they're bankrupt, and Wikidata did put one in because of my book, but it hasn't stayed in there, so that's really annoying. So now I've got to have a look at that one again and figure things out. But yeah, I've gone off on tangents, but disambiguation is very important.

​

Mark A Preston: So, regarding entity SEO, how does Google and the whole ecosystem look at things that have multiple meanings, like "viola"? It's a flower and an instrument. If you are selling, say, the instrument, then what things do you need to do to come up for that?

​

Dixon Jones: I think, rather than jumping on and saying what Google is doing, where I might be right and I might be lying, what I can do is say, well, how does Inlinks do it? Because Inlinks has its own natural language processing algorithm. So we look at pages and turn content into knowledge graphs, and that's kind of what we do. And so we spend a lot of time trying to work out whether this page that talks about a Mustang is about a car or whether it's about a horse or whether it's about an airplane, for example, because there's an airplane in the wall. For us, it's proximity to other entities. So if you talk about Mustang and tires and car, then these concepts clearly tie it to Mustang the car. If you talk about the wilderness and stables, then probably you're talking about the horse. The problem is you've got to be a little bit careful with the algorithm. The algorithm has to be a little bit careful because things like horsepower could apply to both concepts really.

 

But it usually doesn't take a lot of entities, particularly entities that appear in H tags, which are the meaningful parts of a web page. These ones for us carry a bit more weight than talking about something deep down in the body of the page as the primary ideas of a page. So when we look at a page, we sit there and try to work out the most important ideas on the web page. We then associate those ideas with a Wikipedia URL, so we know what entity we're talking about. That's very important because we know that Google also used Wikipedia. Freebase worked on Wikipedia. Wikipedia is an open-source dataset that anybody can work on, but we can also use it to translate ideas. So if I sit there and say, "This page is about this Wikipedia article," then Google's Knowledge graph will have a database of ideas of anything. But one of those will be, "What is the Wikipedia article that refers to it?" So it allows a two-way communication, not just with Google's database, but with any serious knowledge graph that's really trying to gather information. So where the faults come is where you kind of get errors in the system, and you do get errors in the system. Within our system, whenever there is an error, if a user sees an error, they can press a button. I'm angry about this button. It just says, "fine, thank you very much." At that point, Genie, who manages our Knowledge Graph, gets notified that somebody was not happy with the results, so we've gamified it into the system. Genie is constantly looking at the knowledge graph.

 

So whenever people are analyzing new websites, that generically will then go into our system to see, "okay, is this a concept that we've come across before?" And so now it's kind of accurate for most people. But if you suddenly come up with rocks in the Paleontophic from the Paleontopic era, and we haven't done anything about that before, then we're going to make errors and we're going to not understand the Wikipedia pages, and someone is going to be saying, "well, that hasn't done a very good job." At which point, Genie can say, "right, we've now got demand in here, we need to work those things out." But essentially, it's a database saying a Mustang is a horse and has four legs and lives outside, whereas a Mustang has wheels and an exhaust pipe, except mine, which is electric. Then, that defines the differences by the things that you're associated with. So that's the difference, really. And it works really well for Google when they kind of decided that looking at the world in terms of keywords and pages just became too big. The world increases by millions of pages every day, probably billions, but it doesn't increase the number of concepts in the world every day. The number of concepts in the world is not growing so fast. And so, organizing the world's information by idea seems immensely sensible for a search engine when it gets to that kind of scale, to Google scale, but it does lose some granularity at that point. You can't sit there and say, well, everything called Dixon Jones is an architect, when there's at least one Dixon Jones that says, hey.

​

Mark A Preston: So when you come up with the idea to create Inlinks, how long was it in development before you launched it?

​

Dixon Jones: Okay, well, firstly, not my idea. Same with Majestic, not my idea. So my only tip in entrepreneurial life is online marketing is: make sure that your developer has more shares in the company than you do, because the easiest thing is for your developer to build something and then piss off. And that's going to happen however much you pay a company to do the development for you. They're only interested in making new stuff. They're not interested in maintaining and developing stuff that wasn't their own idea. In this case, InLinks is Fred Laurent. So after I pulled back from Majestic, I took some time off just to recover from what was quite a grueling decade, really, I think a hard work decade of running around the world, going to conferences and stuff. Took a bit of time off and then I changed my website@dixonjones.com to reposition myself as I'm open to be on board, on boards, be a non-exec director or possibly invest in your business, that kind of thing. But please fill in this web form. What I didn't want is people to come out of the blue, just ask and just drop things.

 

So the form was a simple form, but it did say, don't do it unless you're a developer, because basically I wanted to help a developer because I didn't want to help another marketing person. I had nothing to add to a marketing person. And then, hopefully, it was a business within the search, at least in the internet marketing industry. And Fred was one of a few people, in fact, I'd probably say only two or three people, properly ever filled in that form. Still up there today, but most people didn't read the form and that annoyed me, whereas Fred filled it all in. He knew who I was, he'd looked me up, he had an idea, and exactly like my relationship with Majestic, really, as soon as the developer had made contact with me, I kind of knew that I knew everybody that probably would want this product. So yes, it was a way off being a product with the product that the public could have, but it was worth me investing in. I put some cash into it. We had a proper joint venture arrangement. I spent probably as much on the lawyer as I did on the investment, and we made sure that the intellectual property was part of the company, that everything was as good as possible, which was kind of hard because he's in France, and it was just as we were all going through Brexit. We actually found it very hard to set up a bank account in the UK.

 

Because he then wanted me to be the CEO rather than the marketing director, which I understand, but it's not necessarily my favorite role. I'd rather be the guy that's out there just, you know, hammering the brand out there. But, you know, somebody's going to look after the books, in which case, we're going to make it a UK company, because I've got no idea about law in France. And I certainly don't want to have any legal foothold in America until I really have to because they'll just sue us because we're not American in my head, whether that's true or not. So we set up a UK company. To this day, Fred hasn't cut. Oh no, he did come in. He came over for Brighton last year, but we kind of meet two times a year, but pretty much all of it was done on Teams before - Teams on chat and email and stuff like that - and we just met a couple of times before we signed the deal, really.

​

Mark A Preston: I noticed when Inlinks first launched, the first few months, I spoke to a lot of people in the SEO industry, and they just didn't get it. They just didn't understand what it was. How did you?

​

Dixon Jones: I think that was inevitable, really. I think that you have to educate the audience. I mean, if you're going to be able to come up with a value proposition with an internet marketing tool at this day, this late in the game, so to speak, then it's going to have to be a good six months ahead of everybody else in its thinking, and this was a long way ahead, I thought, but we had to end up having to write the book just to get it into my own mind as well. But also, the content of that book was also us developing a knowledge base on the website for everybody to eat up and start to understand. But we used the phrase entities, and Andres from Word Lift used the phrase semantic SEO, or Bill Slowsky used semantic SEO, so we're using different phrases, I think. Jason Barnard was all about entities as well, but he thought about entities in terms of you as a brand or your business as a brand and how you could then soak up everything around that. So he was thinking about the brand out.

 

So there's three or four of us were really thinking about it. I'd say Word Lift were making a business out of it. InLinks were... there's a company called TextRazor which is a natural language processing algorithm from France, but not a huge amount that we're really investing in the concept. But I didn't feel that  mattered because we were building something which we wanted to make sure that it wasn't perfect yet. So you kind of need the forerunners, the people that are willing to take that risk and dive in and understand the new stuff to help you shape it but also help you explain it. Getting feedback from a small group of dedicated individuals was worth the wait really because it stopped us making bigger errors or the same errors later in the day with a bigger audience. Because when you make an error, that gets amplified on Twitter much quicker than when you do something good. So it's good to go a little slower. But, at the same time, I still think InLinks hasn't got a huge leap in its brand recognition yet. But, of course, we had COVID as soon as we went into everything. So my modus operandi for reception and for Majestic was to go to conferences, speak at conferences, drop my name in, drop my brand in these kind of ideas. You know, conferences were a major marketing channel for me and my methodology, and then closing them all down. Yes, on the plus side, it cost you a lot less to go and do a zoom call.

 

But, on the downside, it's harder to really engage with somebody. It's easier when you're face-to-face, well, for me anyway, the trust seems to build up better in both directions really. Some people just pretend to be watching if it's on a screen and who knows, the screen could be off and it could be doing all sorts of things. It's either in a room, you're either in the room and watching and engaging or you're not in the room and so it becomes a much smaller organ group of people, but they can get it a lot closer. And you've got to get every single one of those in a SaaS model. Every single one of those people that signs up for a free account or a paid account or whatever, you've got to get over the line. They've got their own story to get there. So you shouldn't underestimate the importance of each of those individual customers. Even though I'm going to need thousands to make it all work at the end of it, each one along the way is important because they're not just a customer, they're a voice. And what they say is important. Not all of them are going to like it, but hopefully enough will.

​

Mark A Preston: So, what are the advantages of using InLinks as opposed to doing what InLinks does, but yourself manually?

​

Dixon Jones: Inlinks at the moment does three things pretty well. It's come out with a social media thing which we can talk about, if you like, more than happy to, which is coming out next week. But the main one, I suppose, the reason it's called InLinks is that it does internal linking. So what it does is allow you to internally connect ideas because it reads your website and turns it into a knowledge graph of ideas. We know what you're talking about on what pages. We then have a human in the loop, the user who sits there and says, "Right, these are my main SEO pages and this one's about this concept, this one's about this concept, this one's about this concept." So they're basically selecting and putting a human in the loop just to stop the AI from going haywire. We tried to do it automatically and it worked sometimes, but not enough to make it automatic in the tool.

 

So we have this human in the loop. And then as soon as you say, "Right, this is the page about whiteboards," then all other mentions of whiteboards will link through. Now, the interesting thing about it is that if you try to do it with a plugin, so link with Spruce a plugin, for example, that goes into WordPress, where you can do it by anchor text and it'll just say, "Right, every time you mention this word, it links through." The problem there is that if you've got a page on towers and a page on bridges, then when somebody says Tower bridge, then that's a very different concept to a tower or a bridge. It's a bit of both, but there's a different concept. So, the advantage of using an NLP to get these ideas is that the anchor text is more spread and hopefully relevant. So, you can say, for example, when you mention towers on a page, I want links to this page to be about bridges, but only when they mention London on the page, for example. So, you can then put context in there and it will do it on the fly and automatically inject it using one line of JavaScript. If you do that manually, then you've got to go through each page on your website. Let's say you've got a modest site, well, let's say you've got a 50-page website. Reading 50 pages and understanding that content is one thing. Reading those 50 pages and understanding all the concepts on each one of those pages, there may be 100 concepts on each page, 20 of which are important to you as pillar pages. So, maybe you got 20 pages where you want the authority to lie in Google's eyes around a concept. You've still got 50 pages to read and you're not reading and looking for one idea. You're reading and thinking and trying to find all 20 of those ideas and where they are. You're going to make mistakes as a human being.

 

The NLP algorithm is going to make mistakes as well, but you can then see them and see where those mistakes are. It's going to make a lot fewer mistakes, and it's going to be a lot, lot quicker. So something that would take forever and never be right, we can do in literally minutes if you connect things up, although you might The NLP algorithm is going to make mistakes as well, but you can then see them and see where those mistakes are. It's going to make a lot fewer mistakes, and it's going to be a lot, lot quicker. So something that would take forever and never be right, we can do in literally minutes if you connect things up, although you might want to edit it. So that's the internal linking. The schema is also extremely hard to do because you can't just read a page. You can read a page and say, "Hey, this is about bookcases." But then you've got to go and look up the Wikipedia URL for bookcases, and it's about bookcases and office and guitar. There's a guitar behind me. These are the things about this page that are important. So it's about this, and then it mentions these concepts. You've got to go and find Wikipedia URLs for all of those to write all the schema for all of them. If you want to do that for 100 pages, that's going to take you a long, long time. So again, it does that very, very quickly. It injects a schema. If you then change a page, for example, it'll rewrite it all. Whereas if you change a page and you've got to go and rewrite the schema, that's going to take you right out of your comfort zone. The content optimization bit is also something you can't easily do. You'd have to read the top ten web pages for any particular phrase on Google, try and build your own knowledge graph of concepts that they're all talking about, maybe put that into an Excel spreadsheet. You certainly can do it, but doing it at scale is very, very hard though. So it gives you that scalability and a level of accuracy that you can't do as a human. So those are the real reasons why using a tool like InLinks just makes it a lot more straightforward and hopefully a lot more accurate.

​

Mark A Preston: So once you've used the tool, and once you've set it up properly, and it's embedded on your website, what sort of impact does it generate?

​

Dixon Jones: It depends a little bit on the website. I mean, if you've got a well optimized site that Google properly understands, then probably the impact is going to be a lot smaller than if you've been writing content and you've just been doing it without any thought to search engine optimization and stuff. A website that has been developed without much thought to search engine optimization is likely to not have thought through ideas very efficiently, certainly won't be connecting ideas properly, hasn't decided which ones are the pillar posts, if you like, of your content from your business point of view or search point of view. And so then the effect is very dramatic to the point of getting that line that goes like, that will happen, but it doesn't happen for everyone and it doesn't happen so much for the sites that were good before. What it does is it disambiguates any errors that Google sees. It stops all the cannibalization because the internal linking explains to a search engine where the authority is for a concept on your website and makes it very, very clear.

 

And so that becomes very valuable. When you're writing new content, though it doesn't matter where you start, it's a very effective tool for writing new content or for going through an existing page and auditing it. So you can go through. If you wanted to audit an existing page, what it does is it's got your page about Mustangs and all the things that you've talked about on there. It'll then go and look at Google to the top ten web pages for Google for the phrase Mustang, or hopefully Mustang horse or Mustang car, whatever you're going to go for, you can change those, but basically that gives you a seed set of the pages that are authoritative around a concept. So that's the idea though. So you can change them. You can put in any ten pages if you wanted to, but it's going to read those and say, right, these are the concepts that are important for this particular phrase, that's kind of a little graph of ideas, and then this is what you're talking about. And you're missing fuel consumption or you're missing V eight or whatever is these important elements are missing from your content. You can't possibly be an authority in this content if you're talking about a Mustang without V eight or whatever it may be.

 

So these are the ideas that really help to improve a page. That's why you and I fight over a particular phrase on search. And I do enjoy it, by the way, you beat me, but there's only so good I can get with the AI content. But you're number one for SEO speaker. I'm number two for SEO speaker as of today anyway. But I use in-links to change my content every time you pop up above me, or every time somebody else comes in second above me, I can go back and have a look and say, right, okay, what's happened? What's changed here? And I can modify my content. It may not just be content, of course, there are other factors that affect that situation, but it's still my major tool for optimizing a page for a phrase.

​

Mark A Preston: So I want to just delve into the Schema side of InLinks a bit more. So, what's the entity SEO and Schema? Is it two different things, or is Schema just part of entity SEO?

​

Dixon Jones: Okay, so the problem is that the language has moved quite quickly, and I don't think there's necessarily an agreed view on the way that everybody looks at things. But the thing about schema is, yes, it is based around entities and concepts, absolutely. But it's much more varied than what we offer on InLinks. InLinks is just about reading a page of content and seeing what the entities are and being able to communicate specifically to say, "This page is about this Wikipedia page over here and this Wikipedia page over here." That's pretty much it. We do some FAQ schema, we'll probably develop more schema in the minute, but most tools are much more about what is your organization. So if you use a Yoast plugin or rank math, they've got schema and stuff in there as well. But it's like put in the name of your organization, where is it located, what date was it formed? Or if it's an individual, when were you born, when did that person die, or who are they married to, whatever it may be. So it's about relationships between ideas. All of those are important, but these are generally site-wide.

 

So if you've got British Airways' website, you would hope that British Airways have some schema on. They're probably not; they're probably too big to think about it, but anyway, that they've got British Airways, and this is the definition of British Airways, and this is a link to the Wikipedia page and stuff like that. But that applies to the whole website, not to the actual content on the page. So, schema is really all about trying to turn content into meaning, out of content in a form that requires a lot less memory and is a lot more specific. So, I kind of describe it as the meta keywords of today, really. You know, it covers so many different areas. You've got schema for recipes, you've got schema for events, you got schema for people, you got schema for types of football teams. I don't know, but all sorts of different things have got schema associated with them. And schema.org just goes on forever and ever, and it's a living document that goes out there. And some of this schema, a lot of the schema, is really powerful for Google to help show rich snippets. It helps it to understand that this is an answer to this question or this is a recipe for this product or whatever. So, using a schema helps to disambiguate whether this is actually a recipe or whether it's a way of building a house. And we're not currently involved in all of that part of it. We're just involved in this little bit, which we think is highly relevant to the original concept of search, which is, I've got this page of content. How do I make sure that when somebody searches about this, that this matches? And we think that schema is very helpful for that purpose, but doesn't actually affect the rich snippets very much, but it can help us to communicate with knowledge graphs.

​

Mark A Preston: Moving away from InLinks slightly, you do a lot of traveling conferences and everything. What do you think changed in the conference world in the industry?

​

Dixon Jones: Well, COVID changed, for sure. I mean, that changed everything. But I think conferences are starting to come back with avenge. We'll see. I mean, the last time you and I met was down in Brighton. That was pretty packed. I remember sitting there chatting to you in Brighton. We were getting bumped from all sides by people trying to walk past, so it was fairly full. I think it was probably not as full as it had been before COVID, but I think that's coming back now. I think that there is a big appetite for people to go and meet people face to face again, and I think that we did what we could with COVID. It wasn't without its advantages. We probably saved the planet by a year by not traveling as much as we all do. But on the other hand, we all went slightly stir crazy at the same time, and that stir craziness is just as important to our mental health as climate change, not that they're both not both critically important. I think it's a fair phrase to say, but I find that ideas spread much better in a face-to-face environment. And if you can get a bunch of people that are passionate about the same thing in the same place, then it's a very enjoyable evening. I wouldn't talk about SEO to my chess club, but also I wouldn't not go to my chess club on Tuesday nights because I really enjoy that chess community.

​

Mark A Preston: So in the speaker world, when it comes to conferences over the past year or so, has it improved than it used to be?

​

Dixon Jones: No, it depends conference by conference. I go to a conference in Slovenia every year, for example, and I'm the MC of that conference, and I thoroughly enjoy it. But it's not an SEO conference. It's a digital marketing conference, for sure, but SEO is a very small part of it, and I find it really interesting because it's quite high level, and I really enjoy it. But at the same time, when I go to a very specific conference, it can be right on point. But I do find that the deeper you go on point, the more the presentations are weak, actually. There's a move to try and be inclusive and to try and get everybody in there. I think that has a tendency to sometimes allow talks to get in that are not really well thought through. Don't get me wrong, I've been responsible for many bad talks in my time, so I'm not suggesting that I don't do some bad talks as well, but I think that the quality of really good ideas are few and far between. The new ones are few and far between. Maybe that's because I'm so old in the tooth that I've just been to a lot of it, but sometimes you can go to presentations and there's a few people in there and you go, "Well, I want to watch them because they've probably got some ideas," but you're lucky if you get there. I think you need to have people with good ideas filtered out by the organizers before you come along so you've got these new ideas and these new concepts being talked about, and maybe you've got a lot less presentations but they're all going to be full of sham packed with good stuff. But I don't think that's a function of things getting better or worse so much as just probably has always been the case, and not necessarily just in our industry. If you're a mechanic and you know how a car works, you don't need to have that story again. I've got no idea how a car works, but I also don't want to learn.

​

Mark A Preston: Funny enough, I am a mechanic and I think a lot of my logical thinking comes from mechanic days when cars you actually could strip down and fix, rather than machine telling you what's wrong, that logical thinking. I've just got a couple more questions. Next one, if you had a magic wand and you could change just one thing in the industry, what would it be?

​

Dixon Jones: Oh, I think I would not have one dominant player in Google. I think that it really undermines an SEO consultant's ability to perform if they have to continually jump through the hoops of one engine. I would much rather have a splinter industry with 50 search engines and many more channels to market than one big player.

​

Mark A Preston: Wonderful. Finally, is there anything that you really want the audience to know that we haven't spoken about yet? 

​

Dixon Jones: No, I think you've been very kind about InLinks and thank you very much for delving into it. If anybody wants to play with InLinks, it's Inlinks.com and you can get a free account for your first 20 pages to play with, and I hope people do. But if you go on the blog, it's got a semantic SEO guide which is largely a big part of the entity SEO book. You don't have to buy it, it's just there. And it was me going through for about 18 months trying to figure out this entity-based approach to search engine optimization, and I hope people read it and give it a try.

​

Mark A Preston: Good. And finally, what conversations do you want to have and where can people find you?

​

Dixon Jones: You can find me at dixonjones.com. I pay for the blue tick on Twitter because I'm not clever enough to get it, since Elon Musk has banned the ability to give it for free. But Dixon underscored Jones on Twitter. Yeah, if you type in Dixon Jones and can't find me, then your searching is very bad.

​

Mark A Preston: Just dot search SEO speaker.

​

Dixon Jones: No, no, don't do that. And don't mention the architects as well.

​

Mark A Preston: Well, many thanks, Dixon, for joining us, and totally appreciate your time.

​

Dixon Jones: No, and yours as well, Mark, and thanks very much, and see you in the SERPs.

​

Mark A Preston: Yeah.