Dixon Jones is a true expert in search engine optimization.
He’s no stranger to changes in SEO, and he recognizes a fundamental shift happening right now – and it’s one that businesses need to understand if they want to make it to the top.
In this podcast with John Lincoln, Dixon Jones discusses what may be the biggest change to ever hit the SEO industry.
Listen to the Full Episode:
Listen on Spotify, Apple Podcasts, or Google Podcasts.
Full Transcript:
John Lincoln: All right, everybody. Welcome to another Ignite visibility University podcast today I have Dixon Jones on the line. I’m really excited to have him here. He’s a respected award winning member of the internet marketing community with 20 years of experience in search marketing and 25 years of business innovation experience, and he’s been connected to some really big projects, one being Majestic, which is one of the cooler linking databases out there.
He’s got an MBA strategic management. He’s a really smart guy and I’m very, very happy to have him on the line today. So Dixon, how you doing today?
Dixon Jones: John I’m doing well, and thank you very much for the kind introduction. That’s great.
John Lincoln: We were just chatting a little bit before jumped it on. Where are you in the world today? You’re over over the pond, correct?
Dixon Jones: Yeah, I’m actually in a little incubator in a place called Cranfield University, which is a fairly well-respected University and the nearest university to my house. So I’ve got myself a little desk here so I can get out of the house. Otherwise the dog kind of shouts at me and that nice calm day doesn’t seem to work that way. I don’t go up to the Majestic offices too much anymore. So it’s just, you know, nice to have a base that’s not the house.
John Lincoln: Makes sense. So for everybody on the podcast, tell us like just a little bit about your entrepreneurial journey.
Where did you start? Where are you today? I know Majestic is part of that. And then you also have some new things going on with inlinks.net, but just a little bit of background. How did you get started and what did your journey look like?
Dixon Jones: This quite a long journey. So I came out of university in 1988-89 and there wasn’t an awful lot of jobs to be had. I actually was president of my Students Union for a year and then set up a business, which was actually writing and running murder mystery evenings for a living. So kind of a bit off the beaten track, but when the Internet came along I built a website and found that people from Ohio are coming to my website and I couldn’t get actors to Ohio. So that’s how I learned about search engine optimization. I put up a sort of a downloadable version of the website and of the games that I was was running, and that’s how I learned about SEO.
Then in 1999 or something like that, I kind of set up a consultancy that was doing search engine optimization. I call it a consultancy, nowadays people call them agencies, but I didn’t know what an agency was. My first ever client, I think, was a subsidiary of Mohammed Al Fayed, who used to run Harrods in the UK, and it kind of went from there.
I sold that agency some years back, but about 10 years ago I started getting involved with Majestic. That was kind of one guy in his bedroom really when I went out there with Alex, who’s the founder there. And together, we’ve built it up with with Steve Pitchford into a reasonably big business and and that’s turned into something that kind of changed my life. And it’s been very successful.
John Lincoln: How many people work at majestic now.
Dixon Jones: Well, I’m not actually allowed to talk about that, not least because I’ve not been there since January, so it’s not as many as you might think. But it’s amazing how many computers there are, truckloads of conditioning. Literally, they sit on the desk on the floors of the offices, and they’re quite big offices, you know, but I’m making computers arrive on trucks to put them in data center. So, you know, an awful lot of computers. I’d say it’s no longer a small company, but it’s certainly not like some of the US companies seem to go, you know, to hundreds of people straight out of the box. We’re much more about the computers than we are about the individuals.
John Lincoln: Excellent. Well, Majestic is one of my favorite link analysis tools and the biggest link database out there online. I’ve seen that touted a couple of times.
So, tell me a little bit about inlinks.net. That’s something you’re working on a bit.
Dixon Jones: Yes, that’s actually in closed beta at the moment, or if you try really hard, you can get on to it.
I pulled back from majestic really 18 months or so ago and I needed to get some space. The guys were moving in a way that was great, but it was probably best if I kind of started again, because I’m really good at getting something off the ground. I’m not so good at the politics of larger organization. So, I kind of like pulled back a little bit and I’m still, I’m the global ambassador for Majestic, but they were working on some new stuff and I wanted to work on some new stuff.
With Inlinks I went in with a French SEO expert, and he has really started to understand semantic SEO, so entity-based search, which is really interesting. It’s really all based around the Knowledge Graph and the Knowledge Boxes that you see on the side of Google and everything, which is pretty interesting. But Google’s got some pretty good API’s behind that and a huge amount of theory behind it.
So, what we started to do was to look at the content on web pages, not based around keywords but about what Google is saying the content on the page is about. And I’m not saying the whole page. I’m saying about every little sentence and section of the web page. Google brings out some information about what those pages are about in snippets.
So, what we’ve done is built a tool that will go and analyze all the pages on any given website, or the top pages on any given website, and find out what all the little minority entities are that are being talked about within that content, and then try and help you associate landing pages with those topics. Then, it will automate the internal linking so that every time you mentioned – for example, Cranfield University – then it links to the University’s page or, if it talks about puppy dogs, then it puts a little link to the breeding page and then that kind of thing.
So it’ll take all the entities that are all over the content of the website and link those up so that Google can pretty well understand which page is supposed to be with which entity. So far, it seems to be working pretty well.
John Lincoln: So let me see if I understand this correctly. Is this an internal linking tool that I would install on my own website?
Dixon Jones: Yeah, it is. So you do have to put a line of JavaScript on the side. The great thing now is that Google has gotten so good with the JavaScript stuff that we can actually start building tools that Google can can eat up and work with, and it saves a lot of that backlink between the SEO agency and the developers.
It’s one line of code with Google Tag Manager, and basically it’ll add internal links onto your website. It’ll also add schemer onto your website. It’ll also analyze web pages and and look at the content from an SEO perspective and say, right, okay. Well, the other pages that are ranking with this kindof phrase or idea are using these ideas and talking about these ideas which is semantically very close to the ones that you’re talking about. So, you know, it builds out content as well. But there’s no external linking involved in this at all. It’s just links within your own website. We didn’t want to go off on the direction of “you’re accused of selling links” or anything.
John Lincoln: So, this is cool. I remember back when I was a director at this other company, part of our program was installing a WordPress plugin which would allow you to set a certain anchor text, which is the HTML within a link for anybody who doesn’t know, and then it would allow you to set a certain URL, and then it would crawl the website and find any time it was mentioned and internally link it. I haven’t used that plugin for a long time. This sounds like a much cooler version of tha, looking at more of an entity and more of an automation. Is that correct?
Dixon Jones: Yeah, because that plugin and that idea was brilliant in its time. It was an excellent idea but as Panda came out and, well, I think it was Panda. But, you know, correct me if I’m wrong.
John Lincoln: Penguin I think.
Dixon Jones: Then they started to really penalize the anchor text and heavy use of anchor texts around specific keywords that became a little bit of a problem, and because the site was fairly obviously internally linking around very specific keywords. But the idea of linking ideas together is really where they want the Knowledge Graph idea to go, so they want the user to be able to easily flow around between ideas.
And if you look at Wikipedia, for example, I was analyzing a particular Wikipedia page today for popcorn actually, and it’s a page with a huge amount of content on it, but it’s got something like 700 links on the web page and it ranks for the big word, as you might imagine, it being Wikipedia and all, but it’s because internal linking is so heavily done on Wikipedia that it’s really helping Google to understand the flow of ideas on Wikipedia. And obviously, you know, Google treats Wikipedia as a trusted data set anyway, but even without the amount of structuring those internal links really do give Wikipedia a lot of power.
So, I think the internal linking is really important, but the anchor text is so much less important than it was before. It’s linking ideas together rather than linking keywords together. But yes, you’re right.
John Lincoln: Yeah, I think a lot of people ignore the power and the importance of internal linking when it comes to SEO.
Dixon Jones: I agee, and honestly I think that people have got very, very lazy and it’s especially true if you’ve got multiple writers on a blog post or a site, then the writer will write their content, they’ve got no particular interest in seeing why a lot of the content is already there that they should be linking to within the same website so they don’t have this urge to do that, and there’s certainly no level of understanding that if they did connect all those dots than everybody would win.
And also, it’s a very hard thing to do. You can’t just go through with every phrase and then look up, you could do a site search, I suppose, for every single phrase on your content and try and find out what other local versions or URLs are discussing, but you’d still then potentially link ideas or multiple authors would link ideas to multiple pages and it still wouldn’t be clear to Google which page is supposed to be the entity about any particular entity or idea. So automating that, or at least semi automating that, I think is going to be really, really powerful. We hope so anyway.
John Lincoln: I love that idea. That’s great. I can’t wait to try it. Make sure to, you know, give me anything I can to to give it a shot.
Dixon Jones: It’s a super secret beta link John, don’t tell anybody.
John Lincoln: I won’t tell anybody except for all the people who are listening.
So, one quick thing, and this is a little bit different. And then I want to move on to a couple more questions I have for you about linking. Just a tip for everybody who’s listening here today: one of the things I love to do is sort a website by the landing pages that have the most xxternal links pointing at them, and then go back through and put in internal links to the most important pages, with some variation of anchor texts that will help them rank better. So, finding the pages with the most links, the most clout, and making sure you’ve got good internal links on them. But doing this type of thing at a mass scale, I just love that idea.
So, Dixon, let me ask you a question. So linking is a big topic and it’s something that you probably know more about then maybe anybody out there, or most people in the world, that’s for sure. So let’s talk a little bit about linking in general. When you look at links for SEO, what makes a good link? What will make a bad link? Just generally for our listeners here who aren’t super savvy at search.
Dixon Jones: OK, so I think I’ll start with what makes a bad link because, I mean, there are a lot of things that make a bad link. Really the worst ones are buying a link and Google figuring out that you bought the link. That’s pretty bad. Or, on the plus side, Google will give you a nice big message in your Google Search Console that you’ve now got penalized. So, if you get caught buying links, Google doesn’t like that.
But I think that the other things about bad links are that it’s not so much that they’re all bad links, it’s that Google can pretty much discount the vast majority of links. I’m talking about links from other websites to your web pages now rather than internal links within your web page. And most people think, Hey, oh, I’ve got a link from, I don’t know, Forbes or Reddit, and that’s going to be great because Reddit’s a really important site and there’s loads and loads of stuff out there. But really it’s just somebody who’s going and putting up user-generated content onto a website, and anybody can do it. And Google figures that out pretty quickly. It’s not that they’re bad, it’s that they’re not going to count for very much in the greater scheme of things.
I think the other thing that people don’t really realize is that there can be lots and lots of links on a web page and yours is only one. And so Google has to split out the – whatever you want to call it, whatever your magic word is – the link juice, PageRank, the value of a web page. It has to decide in some way how it’s going to spread that equity. Certainly as a page that has only 10 links on it is going to give out proportionally much more love to your site than if it’s linking to you from a page with 1000 links on it.
And the third thing is that I think a good link is one that is in the body of the text above the fold, and naturally there and helping the user get more information about the subject that they’re currently reading. So you don’t really want a link that is so completely out of context that the user might not travel down.
And then how you imagine a machine might analyze that is a whole new thing, and Majestic’s come out with some stuff that tries to help that, which we may not have time to talk about now, but happy to if you wanted to.
John Lincoln: So the best links, and I agree with you, the best links are in-content links on a highly authoritative web page that’s been around for a while and that has been linked to from other websites out there that have also had great links and it’s been built up over time. I mean, you generally agree that that’s maybe not the best, but that’s a good link, right?
Dixon Jones: I’d say that’s good and I think you can either get mathematical about it or you can say yeah, if a user would naturally go down that link, then that’s probably a good sign. But the mathematics is quite interesting because, you know, above the fold is probably better than below the fold, and in the body text is probably better than not body text. A follow link is probably better than a nofollow link, you know, and there’s a whole load of stuff that you know can raise the visibility of that link based on how many other links are on the page. How many are internal, how many are external and all the other bits and pieces.
But you’re right. It’s the power of that page and the context of the link.
John Lincoln: Alright, awesome. I want to try a little exercise with you. It’s a little rapid-fire.
Is this a good link or a bad link? We’ve got a global external link on a website that goes from one website to another website. It’s a coupon site with the anchor text coupon.com it’s on every page of a 100,000 page website, going from one site to another site. Would you want that link or not want that link?
Dixon Jones: Right. Okay. There’s a whole load of things about coupons right now.
Well, firstly, site-wide links to other sites have been, I think Google picks those up and says, “right, a site wide link, I’m not going to count as a site wide link.” So if it’s in the footer of every single web page, then I think that Google is going to say, okay, that’s not 100,000 links that’s, you know, at best one link.
But I think that they have also said that they’re going to penalize coupon sites. But the interesting thing is what they’ve done is they’re penalizing coupon sites that are in subdomains of big web domain. So there are some companies out there, I’m not going to name and shame, but they’ve done a brilliant job up until now. But there are companies out there that I’ve taken subdomains of really famous websites. And so that’s where they put all the coupons and things for all of the promotions and things that those those websites do.
So, let’s try and do one that I don’t think is involved. So let’s say Wikipedia wanted to monetize their website and they might link to discount.wikipedia.org, and that discounts.wikipedia.org then contains the nuts and bolts of voucher coupon codes. And Google, I think, has decided that kind of activity. They’re trying to penalize because the coupon company is then owning the sub domain and trying to hide under the cover of the big brand, and they’re getting a lot of links into that from the main domain. So I think that’s something they’re targeting right now. Site-wide link is fine, but I don’t think it’s going to count as a hundred thousand links.
John Lincoln: Yeah, those little coupon sites. I’ve worked in that space actually just a ton, and then the global links, I think one thing I’ve seen a lot of people do in the past is, you know, say I own an agency. So every website we made, we put a global footer links back to our website with the term “digital marketing agency”
Dixon Jones: Seen a lot of that before.
John Lincoln: Right. Yeah, lots of people did because it was working really, really well for for a long time, really until about three or four years ago. And then all those people got penalties and they had to rebuild.
Okay, so what about directory links directory websites? That was really big four or five years ago.
Dixon Jones: They’re terrible. Complete disaster, and the reason that they’re a real disaster is that there’s only external links on a web page and it becomes fairly obvious: a list of external links and there’s more external links than internal links. I think these stand out pretty highly for two reasons. Mathematically, it just divides all of the love out completely. So you don’t get very much love anyway, but also Google realizes there’s a distinct lack of content. So there’s this other link density that Majestic’s trying to do, they’ve broken down every web page on the internet into 40 different sections. And then they said, right, there’s two and a half percentile. If your link is in there, how many other links are also in that percentile? And then how many actual words are in that section as well? So your link density, a link density of 90% might mean that 90% of all the text in that two and a half percentile is in links and only 2% is actually characters that are not linked. So the link density is very, very high, which is very, very bad.
So, the link density on directory sites tends to be really, really high. Now, if the directory is really, really powerful, then I think it is powerful in its own right. So a good example of that would be Airbnb or Booking.com because they are essentially directories of properties. That’s all they are. But there’s a lot of content on all of those pages and they don’t put them just in a link, they put them with quite a lot of text around those results and they have whole pages of content about each one as well. And then, you know, they internal link in there as well. So you load a page. I’ve got a I’ve got a holiday.
So I’ve got a holiday property in Snowdonia. For example, you know, if it’s in the listings of Booking.com, you know, and comes to that page, then it’s also got content about ballerina link to the banner page or Snowdonia in the Snowdonia page. And so it’s a rich experience for the user in its own right. So a directory has to be good to not be banned by Google.
John Lincoln: Absolutely. Yeah, I agree 100%. One thing we get a lot is, you know, we might get an A prospect and then they’re weighing us and what we’re providing versus the competitor based off of the amount of links. And in the end, I think that the amount of links is not a good thing. It’s about the quality, because we’ll get links based off pure press by doing an industry study, by doing unliked mentions where you reach out to people who have been mentioned online and ask them to link to the website, all this white hat stuff. But a lot of people look at it more as a quantity thing.
My question for you is, for the normal business owner, you know, maybe they have a business that’s 10 million maybe 100 million, it’s a good sized business and they want to build links. What are some good ways to go about it? What are some success stories that you’ve seen? How would you approach it if we’re your own website?
Dixon Jones: Okay. Well, firstly, I absolutely agree with you that it’s quality, not quantity, and it’s even stronger than a logarithmic scale, you know that the strongest links. I think that with link building structures, the more varied your link building structure, the more based around helping users to become more informed the better.
And so I what I mean by this is if you’re going to write things like good insights and good knowledge bits, bits of really good stuff that will help a user to achieve a problem or to understand a problem, then I think that is something that encourages people to link to because there’s no other proper source of salutes or solving that problem. So for example, I wrote a blog post recently on the Knowledge Graph and human bias within it. Most of Google’s Knowledge Graph is all about extrapolation and learning from what Wikipedia does. So people then go to Wikipedia and write content. It turns out that, less than one in 2000 I think it is of people that use Wikipedia actually change anything on Wikipedia. And let’s face it, anybody that uses Wikipedia is a pretty unbiased sample of society, anyway. And so, so that makes the whole of Wikipedia, and I find that quite interesting. I found that a really interesting topic and nobody had really written about it in public. They’d written about it in research papers, which of course I cited a link to. But, you know, it got a reasonable amount take up pretty quickly.
And I think that if you can write a study like that, then people will link to that study because they want to cite what they see as quality, and I cited all the research papers and things I looked at when writing it. So, it’s got to be quality and something that people can reference. And I think that makes for a good campaign. Does that answer your question?
John Lincoln: It really does. And I think that’s a really good takeaway for all the Ignite visibility University listeners right now. Link building should be a content-first driven approach, like creating this amazing asset on your website just like Dixon did and then it’s fine to email people and tell them about it and share it on social, but have it all come in naturally. A nd if you don’t have any great thing on your site, you don’t have any reason for people to link to you, you’re not engaged, even if you don’t have great content on your website in the press and you’re not you’re not getting out there then the links aren’t going to come in.
What’s not good is trying to force it by asking people to link to you when there’s no reason to. So I totally agree with your assessment.
Dixon Jones: Yeah, yeah.
237
00:29:48.120 –> 00:29:51.450
I found an interesting thing about that if we got time for it, but yeah, if you run a thin content site, and that includes a lot of directories, actually, then the interesting thing is that you might run up with a directory structure, which was quite common in the past. Let’s say you sell green widgets and you sell green widgets in California/Los Angeles/Hollywood. And then you also sell green widgets and California/Los Angeles/Beverly Hills. And then you end up with all of those URLs trying to map out pretty much everywhere in America.
But then don’t have any content on all of those pages that relate to each other. So you then can’t link all those pages together and semantically, and then if you can’t do your own internal linking then you haven’t got any decent content on the site. So if you can’t do it, then nobody else in the world is going to link to your content. So yeah, you’ve got to have good content for those links to happen.
John Lincoln: I like that. You brought up one other thing earlier that I want to chat about today. One of the things I don’t think a lot of people realize when it comes to back linking, when it comes to SEO, when it comes to computers and algorithms and all this stuff is that it really all comes down to computing and segmentation. Even what you built with Majestic and Inlinks, everything gets categorized and segmented on some level and then drilled down further and further and further based off of the way that you structure your data analysis. Talk to us just a little bit about that. When you are working on Majestic or Inlinks, what goes through your head, and what goes through the search engines algorithm when they’re analyzing these things?
Dixon Jones: Okay, the thing that 10 years ago attracted me to majestic was the sheer scale and audacity of what they were trying to achieve. The only way to find link information 10 years ago was a system called Yahoo Site Explorer. And Yahoo Site Explorer had really been set up by a guy called Tim Mayra Yahoo who was pretty much trying to undermine Google because they knew that getting backlinks is such a big part of Google’s thing. And we’d done some calculations and thought, that’s costing them millions and they’re not they’re not charging for it. Moz were coming on with the same kind of idea with their Open Site Explorer at the same time, and actually got to market before us.
But anyway, so I wasn’t even with Majestic, and I just found this new tool and I put in, you know, 50 bucks or whatever it was and try to buy some backlink data. It didn’t work. So as SEOs do I lost my bottle and sent rude emails to this company, not knowing whether they’re big or small. And I got this email back from Alex, who’s the founder at Majestic, because he was the only one there. He was still in his bedroom saying, oh, that’s right, we got a bug. I fixed it. Try it again. Here’s some more credits.
And I did it again when I realized that he was the only person I could see that wasn’t using Yahoo Site Explorer backlinks and they actually crawled the internet to do it from his bedroom, more or less. It was a bit bigger than that. I thought, I know everybody in the SEO world needs what you’ve got, so we got together and built out Majestic
And I think it’s the audacity of scale, which is hard to copy. Somewhere during that 10 years someone on freelancer.com, one of those low desk sites, said I want to build something like Majestic on a budget $500 or something. And I thought, that’s really quite entertaining, because it’s just impossible to do that. You’re going to need to crawl the whole internet every day to be able to produce something like Majestic or Open Site Explorer or Link Tool as well, the other major three. And I think that’s the same with links. What I like about that is that it’s okay, we’re not building a data center that large but we’re really leveraging big data into what Fred calls smart data. The algorithms won’t work unless they’ve got a big, big picture and having a subset of all the data doesn’t work so well. If you try to run PageRank maths just on Wikipedia, a guy in Toulouse university did, just because that’s open-source data. He did that PageRank algorithm just on Wikipedia data and found that according to that, Colin Haas the botanist was more influential and more well known in the world than either Jesus Christ or Hitler.
And so if you run without a big enough data set or the wrong training set, then you’re going to get different data with algorithms and that machine learning is something I think is going to be very, very interesting to see how that pans out over the next few years as Google and other search engines and other technologies start to do machine learning, based on seed sets of data, which may or may not be representative of society or the world as a whole.
John Lincoln: That’s a really interesting topic. We’re seeing machine learning come into play really heavy on the advertising side, on the programmatic side, and then definitely search engines. I think the main takeaway is create great content, create a good brand, promote it. Put in the work writing on novel topics. I talked to somebody on the podcast last week about newsjacking so, which is getting involved in the news and being a part of it.
So Dixon, amazing stuff today. I know our listeners are going to absolutely love this. We have a lot of marketers, Chief Marketing Officers, business owners and then SEO people so perfect topic.
I know we’re both speaking at PubCon next week, so I can’t wait to see you there. Other than that, what else exciting is going on in your life right now? Where can people find out more about you and follow you online?
Dixon Jones: You can find me either at DixonJones.com, or my Twitter handle is @Dixon_Jones.
Obviously Inlinks is kind of at the forefront of my mind right now, because we only announced that we had something a week or so ago. You can find me via Majestic, but really Dixonjones.com, and I’ve got an events page of where I’m going to be as well on there so you can always find out whether I’m going to be at PubCon or SMX, whatever it may be.
John Lincoln: Awesome. Well Dixon thanks for taking the time. Everybody once again that’s inlinks.net. I’m actually looking at the website right now it looks pretty cool. By the way, I was not paid or anything to say that I just love cool technology. I actually didn’t even know we’re going to be talking about inlinks.net before we jumped on this podcast but it does look cool. And Dixon, thanks again for your time today and I’ll see you next week. Have a good one.
Dixon Jones: That’s great. Thanks, John.
