Pick His Brain!
I’d like to introduce one of our members, Dixon Jones, for our next ‘Pick His Brain’ session and I want to thank him for the participation.
Dixon has been involved in the industry since 1998. After building up and later selling his agency of 20 people,
Dixon pioneered link analysis tools as CMO of Majestic.com which ultimately took him to Buckingham Palace for two Queens Awards.
Dixon is now the CEO of the new startup, inLinks a Semantic SEO tool that is receiving a lot of attention from experienced SEOs.
Here is Dixon’s Ted Talk.
You can test-drive inLinks.
If you have any questions related to SEO, agency, SaaS, and business development, please pick his brain.
Here are the rules.
1) I’ll let the thread go on until he asks me to stop. Theoretically, this thread can continue until the FaceBook stock value goes to zero.
2) Please, no snarky remarks. I will not tolerate any intentional negativity. We are here to learn from each other’s successes and strategies.
3) Please do not PM him and bother him. If you have a private question, ask for his permission on this thread when appropriate.
Please subscribe to our new Twitter account for the latest updates.
What advice would you give to a person starting a career in Digital Marketing? Tech side definitely.
Great! Tech side is way cooler. I try and go into business with people that are technical and make sure they own more of the company than me… that way I won’t lose my developers!
So on that front, I would recommend that you start to focus on building digital marketing tools that other people are willing to pay for… preferably using a SAAS model.
In the GoldRush, the people that made the money were the ones selling the shovels and pickaxes… not the gold diggers.
Also – if you are just starting out – then maybe don’t do the things I did… they may not work anymore. In fact, the concept of a “website” is falling apart.
API integrations are the way to go. Master Zapier.com and imagine all the ways you could charge people for data. Don’t start out trying to be big… start out trying to be different.
What role will AI and quantum computing play in digital marketing in the next 5 years? And where do you see the human element fitting in to it all. Bonus question: Are we living in a simulation?
“Are we all living in a simulation” Does it matter either way to the decisions you make?
On the main question, I would like to argue that the original PageRank algorithm was a form of AI (more specifically, a form of Machine learning) as it was an iterative model that adapted to its learning ground (web pages) as the web grew.
It was very successful (and based on links). So in a way, the newfound thing about AI being the new “thing” is a bit of window dressing by marketers.
BUT – the change is real. Yes I do think AI will play a larger part. We’ll have to see about Quantum as the technology is not really accessible yet to most of us.
But Quantum computers work on probabilities, not absolutes and (for marketing, which is not black and white) that is a very interesting approach.
There has to be a human in the loop though – because Humans will break machine learning patterns and introduce innovation into the datasets.
Humans cause bias in the initial training sets fro AI, but also have the power to change the underlying model or data.
ASo they remain important. Garry Kasparov wrote a book called “Deep Thinking” which has an interesting perspective from the guy that defended humans against computers for years.
Well worth the read and not too heavy (although, I am a chess fan to the point of seeing him play live once upon a time.)
How do you get around the “how is that page ranking without optimized content”, kinda questions?
You could argue that everything is optimized for SOMETHING.
I am lucky enough to have been out of the agency game for a while now, but if I was agency side again now, I would ask… “How many hours would you like to pay me for to look into it? If that’s just going to annoy the person asking.
I would point them to a post by someone I trust with the Top 10 reasons.
For example, this post: : https://moz.com/…/how-garbage-ranks-in-the-serps-a-case… (I haven’t read it… but if it isn’t good, then someone shoudl write something better!)
Well for newcomer in this industry where I should focus on this filed? And which think will be game changer in next decade.
I would say – play to your strengths and what you enjoy. There’s truth in the saying “If your work is fun, you never have to work a day in your life.”
But I think the by the end of the decade (given that we are 6 days into it) is an interesting event horizon.
10 years from now, we WILL all be heavily into VR and even more relevant in search: Augmented Reality.
I cannot imagine many of us not wearing fancy specs that augment what we see with a data layer.
How cool is that going to be? So – when you look at a building, you will see the GMB data.
When you look at an artwork, you’ll get data from the web (Wikipedia will be discredited by then, because we’ll all have decided the editors are too biased).
Where will Google/Facebook/any other clever glasses look after that?
Start with looking at what Google Maps is doing with this: https://www.youtube.com/watch?v=QW1QT7DOOdA then imagine what happens when the directions turn into passive information?
That looks like an advertising opportunity that Google and Facebook would kill for.
What are your favourite tools and approaches when doing your research? I suggest you use Majestic and Inlinks, but in SEO there is no one tool that does all, what niche tools do you like we might have never heard about? Also in a workflow are there special things you consider? I can easily complicate things as I start to harvest a lot of data and I am looking into revamping my workflow and output. Clients have dificulty with datasets sometimes and yielding through them; when it’s nicely visualized it tend to digest better which leads to screenshot hell (and I am looking to visualize and automate more) but offcourse SEO is different on every site … just curious.
Oh man… my “tech stack” goes on forever.Common ones are SEMRush, Screaming Frog, Trello (I HATE Slack… kill it now!) and of course as you say, Majestic and now inLinks.
But I also like using Startpage.com and a VPN (I use Private Tunnel) and multiple browsers to try to see SERPs “anonymously” or from my user’s location.
I use Excel a LOT and I also use rules in my Outlook to help me filter out leads and data points.
I have been through SO many CRMs and social media comms tools it is not true.
My current gotos are Sendible (over QuickSprout or Buffer) for social and I now use GoSquared for the CRM for inLinks, but I like Agilecrm too.
None of them are good at connecting 1-1 communications across multiple mediums, though, but I have a workaround with GoSquared now.
Looking at my PC right now, I also have, SiteBulb, A bot emulator (although Screaming from does that too, as does a tool by Aaron Wall of SEOTools (old school).
If you want one place to try, though, look at https://freetools.webmasterworld.com/ for a bunch of long-forgotten hacks that still work today.
I hear you on Screenshot hell, but I LOVE “Snipping Tool” on PC. It is built-in, but nobody seems to use it. I have it on my taskbar and use it every day for screenshots.
(MACs are easier on this front).
The workflow thing for clients is hard… another reason why I went tool-side over a decade ago!
A multi-sheet Excel with a nice summary on the front and drill-downs when you need was as good as I ever got.
Oh – and logfiles. Analyze logfiles. I do this in cPanel.
PBNs… still effective or not?
Talk to someone that uses them.
For me, I always focussed on building a business that would last and stand up to scrutiny.
Whether I succeeded or not is for others to judge, but I always found PBNs to not fit in with my approach, which stemmed from theories in my management degree in 1988.
Obviously, I had to adapt, but I dropped PBNs way before Penguin update (2012).
In fact, as soon as I was able to see a measure of a link value with the AC Rank from Majestic (These days it is Citation Flow) around 2009 I realized that directory listings and many types of links sucked.
A good PBN must (one assumes) still work a treat, but unless you are confident of the authority behind those pages, you will always be on thin ice.
One reason I went from Agency side to Tool side was to get away from that “impending doom” feeling that any day.
Google could cut your balls off and leave you to explain an algorithm update to angry customers. I hated that!
Is there an actual systematic/strategic/algorithmic difference between inlinks and apps such as POP, Cora, Surfer etc, or does it do about the same just in a different interface or such?
I’ll stick my neck out here and say there is a core difference… unless those other tools have built their own Knowledge Base, which is possible, but I don’t think so.
(Wordlift does, though… would love to know how we compare to them.)
Almost every tool until recently was based around keyword data, not entity data.
It’s a big subject and I tried to crack the back of it with this: https://inlinks.net/p/semantic-seo-guide/
But unless you are prepared to read “Entity Oriented Search” by Balog (which is free on PDF btw) then it is not so easy to explain the difference.
Also, we are all just getting the hang of Schema (even though we probably should have started 5 years ago) so things are changing fast.
We should probably talk more about the size of our knowledge base, but right now we just want to show people the approach works better without annoying Google too much.
With the schema, it all seems to links to wiki pages. Is there a reason for this? I had an idea of using the ‘about’ and ‘mentions’ to link to Google’s own properties – E.g. if you’re talking about a town then add a Google maps link into the schema for that town – opposed to the wiki page for that town. I’d be interested to hear your thoughts on this?
I have liberally linked to Google properties in the past, but it will not do the same thing as the Wikipedia links in the schema.
Basically (and here my business partner, Fred Laurent might have to correct me) Google’s own knowledge graph stems from Wikidata as a principal seed set.
Infact, Wikidata is the “Human in the loop” that most AI systems need to learn from. It therefore makes perfect sense to reference Wikipedia/Wikidata.
What if there exists a KG for an entity by Google and the entity is not in Wikipedia/Wikidata ? Then would it not be better to reference Google’s MREID instead of having none?
We are looking at ways to give users more flexibility to do that in inlinks, but at the moment we are sticking with what is working… but certainly you make a good point. We need someone to test it!
Would you be interested in doing short educational videos on a Roku channel that is dedicated to Digital Marketing and all the aspects and niches I can add to it?
I’ll accept the friend request – But Roku is new to me. Why don’t people just use Youtube? (…Offtopic discussion to be continued) are you planning on offering an affiliate option for InLinks for people to promote this, at all?
My plan was to use the same idea that Dropbox used… instead of offering hard cash for recommendations, we give users more allowance on their free account usage.
But… whichever way we go… we’ll have to build it first and right now we are working on the two priorities of feature requests (from our few paying customers) and Case Studies (so we can give everyone more confidence in the approach).
Have you seen anybody manipulate google into changing it’s understanding of an entity and/or the semantic relationships that establish the entity?
Good question! Yes.
Understanding more about this day-by-day. Not enough for me to get my name in the knowledge graph though!
So I still have much to learn.
Interlinking within the website should be keyword or entities? Is machine id’s are the easiest way to inform Google about the page content, what we are , where we are from?
Thanks for the kind words.
Q1: yes – connect #ThingsNotStrings.
On question 2, I am not sure if that came out correctly? I can’t see how a machine ID affects links at all?
Although I have seen CDNs proving havoc for some link analysis systems and inflating link counts.
Who would you say helped you most in getting started in your career and what advice would you pass on to people wanting to emulate your success?
When I ran my Agency, I teamed up with a guy 15 years my senior, who I met on a two-day course where the Government were trying to convince Maths graduates to become teachers.
He became my number 2. I owned 40%, he owned 30%, My accountant owned 15% and my (then) Marketing Director owned 15%.
Having business partners really helps and whilst we had our ups and downs.
I think it is fair to say that doing it all on your own is pretty scary.
Also – people like Brett Tabke and Chris Sherman (Pubcon and Search Engine Strategies in the day) let me speak at conferences.
That is also scary, but it makes you test your own beliefs, because they’ll be shot down pretty quick if you bullshit too much on stage (although Future-Guessing is fun).
The early pioneers in SEO mainly communicated through sites like Jimworld (which doesn’t exist now) and Webmasterworld (which does). I used them a lot, back in the day.
Actually, Search was not my first career… I set up a business from University (after a year as SU President) writing and running Murder Mystery Evenings. That started going OK when I realized I needed a partner as well.
Do you think the number of relevant entities in the page is a ranking factor?
Not thought about it. Not directly, but I think indirectly… as long as the entities are connected logically.
Talking about too many disparate thing on the same page might be optimizing for SOMETHING, but probably not for something meaningful.
What are your thoughts on AMP? Ive heard positive things (if used for news sites) and negative news where it decreased rankings/ exposure. Is it a worthwhile exercise (if done right) to increase exposure to sites or is this not a big factor nor worth the time?
I am not am AMP expert. But at the Google Conference in Zurich, Johnmu took a straw poll of those in favour of AMP and those against.
It was evenly split, with a third (including me) not voting. I am sure AMP done right is a big win for many sites, but I only say that from other people’s experiences. I need to channel my time!
What resource(s) would you recommend as your go-to for learning advanced SEO techniques and staying up-to-date on algorithm changes?
By its nature, staying up to date from just one or two sources is very hard.
One person can make headway (and choose to talk about it) but then the next breakthrough or research is from a totally different source.
So… I have Barry Schwarz’s daily roundup hit my inbox every day and I look at the headlines and select what I read.
I keep an eye on Sparktoro’s free citation checker to see what’s current and I go to conferences.
Conferences are really good, because sometimes you don’t want to talk online.
Oh – and I keep an eye on what Google is saying officially as well on their forums, Youtube channel etc.
I do enjoy anything Bill Slawski says – although his biggest posts are about granted patents, which is very interesting but not always actionable.
How those have changed provides an idea of whether or not a patent is likely in use, and how the process it describes has likely been implemented. I’ve been taking some actions based upon those.
What you do is incredibly valuable to the community and, because it is based on something tangible from Google, and because you were a lawyer in a past life, it really has weight.
What it doesn’t have (I am sure you will agree) is timeliness!
Patents take so long to become public. Even so – what you do with patents is brilliant and nobody does it quite like you.
What is the value of a topic / word if Google has no understanding of it?
If you go to the Google Trends tool you can see that Google still happily looks at topics they don’t understand.
They just call them “Search Terms”. Worth having a play, actually. See the screenshot (Kudos to Bill and Bill for being more famous than me in Google’s eyes.
They both paid attention to Freebase when it was accessible. I have more work to do…
Does Google has any anti-SEO algos in place based on your experience? If so, can they be reversed?I see it too often with new websites with new links, those links might not initially help the rankings, but after 4-6 months, we’ll see the impressions rising, also in my experience English websites take a lot longer to rank than other languages, can you confirm that?
I find it very dangerous as a non-googler to say Yes or no to specific algos.
But at the same time, I hate the generic “it depends” response most SEOs give, unles context is offered… so here goes….
Q1: Does Google has any anti-SEO algos in place based on your experience?
Q2: If so, can they be reversed?
A2: Yes. There! I did it! On what you are seeing though, I am not convinced you are seeing anti-SEO algos at play. Links are misrepresented.
Ever since Penguin, most don’t do very much. In fact, the combination of the PageRank maths and the Reasonable Surfer algorithm (and patent) guarantees that a link has to be very good to move the needle.
Most people don’t really understand the implications of either concept. Leave alone understanding the new world of entities.
Your 4-6 weeks could simply be Google taking time to revisit pages that have the links, which it may need to do multiple times for those pages to start building value.
Not all links are created equal. if you use Majestic, I would recommend revisiting the “Link Context” stuff they brought out last summer.
It totally shows why some links are frankly pointless and at the same time shows how Google might do that maths at scale.