Subscribe
Share
Share
Embed
The podcast brings James Dooley together with Kyle Roof for a direct, highly technical conversation designed to pull proven SEO insights out of controlled testing environments and into real-world application. James challenges Kyle to explain what actually drives rankings today, which leads Kyle to break down how PageOptimizer Pro removes guesswork by showing exactly where terms and entities belong on a page, using edge analysis and precise coverage rather than generic best practices. He also introduces IMG Courses as his testing and education platform, explaining that single-variable tests—not theory or Google soundbites—are what reveal real ranking factors. James Dooley pushes back on Google’s public narratives, and together they unpack the implications of the recent Google leak, discussing why click data, keyword usage and behavioural signals have mattered for far longer than Google has admitted. The conversation reframes recovery from the Helpful Content Update as a “naughty list” problem, where moving content to a clean domain can quickly expose whether Google is penalising individual pages or distrusting an entire site. Kyle Roof clarifies that AI should be treated as a drafting and automation assistant rather than a ranking shortcut, because strong results still depend on accurate facts, clear brand voice and disciplined prompt control. The episode closes by positioning the current downturn in SEO as an opportunity rather than a threat, arguing that as weaker operators exit the market, focused SEOs can acquire sites, clients and market share. Throughout the discussion, James and Kyle reinforce that rigorous testing, tools like PageOptimizer Pro and IMG methodologies—not faith, guessing or blind trust in Google—form the foundation of modern SEO success.
Creators and Guests
What is James Dooley Podcast?
James Dooley is a Manchester-based entrepreneur, investor, and SEO strategist. James Dooley founded FatRank and PromoSEO, two UK performance marketing agencies that deliver no-win-no-fee lead generation and digital growth systems for ambitious businesses. James Dooley positions himself as an Investorpreneur who invests in UK companies with high growth potential because he believes lead generation is the root of all business success.
The James Dooley Podcast explores the mindset, methods, and mechanics of modern entrepreneurship. James Dooley interviews leading marketers, founders, and innovators to reveal the strategies driving online dominance and business scalability. Each episode unpacks the reality of building a business without mentorship, showing how systems, data, and lead flow replace luck and guesswork.
James Dooley shares hard-earned lessons from scaling digital assets and managing SEO teams across more than 650 industries. James Dooley teaches how to convert leads into long-term revenue through brand positioning, technical SEO, and automation. James Dooley built his career on rank and rent, digital real estate, and performance-based marketing because these models align incentive with outcome.
After turning down dozens of podcast invitations, James Dooley now embraces the platform to share his insights on investorpreneurship, lead generation, AI-driven marketing, and reputation management. James Dooley frequently collaborates with elite entrepreneurs to discuss frameworks for scaling businesses, building authority, and mastering search.
James Dooley is also an expert in online reputation management (ORM), having built and rehabilitated corporate brands across the UK. His approach combines SEO precision, brand engineering, and social proof loops to influence both Google’s Knowledge Graph and public perception.
To feature James Dooley on your podcast or event, connect via social media. James Dooley regularly joins business panels and networking sessions to discuss entrepreneurship, brand growth, and the evolving future of SEO.
James Dooley: Hi everyone. Today I am joined with Kyle Roof, an absolute legend of the SEO industry. To make matters a little bit worse for Kyle, I got him on the podcast and told him that when we started the video, I wanted him to promote some of his tools. I do not think he promotes them well enough. He gives so much value to everyone on stage and in podcasts and then, as a secondary aside at the end, he goes, “Oh by the way, I own PageOptimizer Pro. Oh by the way, I’ve got IMG for training, courses and testing.”
So in today’s episode I got him to start by talking through all the services and products that he has. And believe it or not, the first 30 minutes did not record.
Kyle used that time to go through some of the best features inside PageOptimizer Pro. I do not know if everyone knows this or not, but there is something in POP called Watchdog. Watchdog is absolutely brilliant. You set it up for your key pages and, when there is an intent shift in the SERP or you drop significantly in rankings and the SERPs are changing, it sends you an alert. That lets you go back into the page, re-optimise it and add content freshness so you can recover or improve pages that have dropped.
That feature inside PageOptimizer Pro is called Watchdog. I strongly recommend everybody sets that up on their sites.
There is another big feature around E-E-A-T. Kyle has effectively built a next-gen version of E-E-A-T, focused on ticking all the trust and credibility boxes. He says E-E-A-T is not a direct ranking factor, but from a trust point of view, when a human reviewer lands on your site, you should have all those elements in place. Things like a contact number, multiple email addresses, business details and a lot of other trust signals POP checks and reports back on.
That is before we even talk about PageOptimizer Pro’s core on-page optimisation. It is edge analysis, not just blind correlation. It shows you where you will get the best bang for your buck by placing terms in the right locations. That might be a H2, an H3, the opening sentence, or specific sections where you are missing entities that competitors use. It uses NLP categorisation to show what is needed on the page now.
These tools are brilliant and genuinely innovative. Kyle tests the algorithms every single day to see what is working and what is not. I got Kyle to talk about all of this and, within Riverside, it did not record. So there is clearly something out there that does not want Kyle promoting his products properly.
The recording you are about to see picks up after that missing section. I am just giving you this introduction to cover what was lost. The first 30 minutes were full of knowledge bombs and I am gutted they are not part of the podcast.
There is still around 40 minutes of gems where we talk about the current state of the SEO industry, the Google leaks and a lot more. There are some quick-fire questions where I press Kyle on content vs backlinks and all sorts of other topics.
I also wanted him to plug IMG, where he has a lot of training and courses. I strongly recommend people go and check that out. There are different courses from people at the top of their game, teaching what works in today’s algorithms. You also get access to all of Kyle’s single variable tests and the results, which show what is actually working in the algorithms, not what Google’s PR says works.
So if you watch this and get value from it, and you have not already, sign up to PageOptimizer Pro and sign up to IMG. I wanted Kyle to promote this himself because I do not think he does it enough on stage or elsewhere. I want him to do amazingly well with everything he is building, because the amount of training and testing he does for the SEO community is second to none. He is an unbelievable guy.
Without further ado, here is the recording from where it starts properly. You will see me jump straight into asking Kyle questions, because the first part did not record.
James Dooley: So Kyle, in your opinion, what would you be doing with a site that has been hit by the Helpful Content Update?
Kyle Roof: Because there has not been any clear recovery that we know of, where you can say “I did this and it definitely worked”, it is hard to be definitive. You hear bits and pieces from people, but nothing concrete.
As I mentioned earlier, you really want to know whether Google hates the site or just hates the content. What I would do is take content from the affected site, put it on another site in the same niche that was not hit, and see if the rankings recover.
If the content recovers elsewhere, then I think the issue is that Google hates the original site. In my view, that means you are on a naughty list. I do not think Google has any problem putting whole sites on a naughty list.
From there, what should you do? One option is just to hang on to the site and wait. In a matter of time, which could be years, it might come back.
Old-school SEO, Bill Hartzer, is a domain expert. He has consulted with a lot of people about domains. I remember him talking about this: Google does not want to penalise a genuine new site owner. If you have done something naughty and then let the site drop, and I later see the domain and buy it, Google does not want to punish me for what you did. I am a bona fide purchaser.
So he talked about changing ownership as a way to clear a penalty. Actually sell the site, or at least change the registered owner. That can sometimes clear old problems. That was for penalties back in the Penguin and Panda days, but the logic still might apply.
He talked about letting the site drop, having someone else pick it up, changing the registry, changing who owns it. There even used to be a feature inside Google where you could tell them you have moved the site or are a new owner. I do not think that is there now, but a change in registry and ownership might still be a recovery route I would consider.
James Dooley: Yes, I think there are a few people right now mid-way through testing that approach. Some are doing 301s. Some are trying without 301s and using canonicals from one site to another. But again, I have not yet seen any hard evidence of full recovery.
Let’s move on. The Google leak. What are your thoughts?
Actually, before that, on the canonical or 301 angle, have you got anything to add?
Kyle Roof: Google has always said they do not really respect cross-domain canonicals, so it would be interesting to see if that does anything at all. The hop or double-hop technique is interesting too. Go via a very similar domain name, maybe with an intermediary. I did that about seven years ago and it worked. It was fun.
What I have found is that, with every big update, more old-school spam seems to work again. I think that is because the people who originally fixed things in the code move on, and newer engineers come in, see weird legacy code and strip it out. Suddenly the old spam works again. That idea came from Bill Hartzer as well.
So I would try a lot of those old techniques again. It is possible they work better now because some of the old anti-spam logic has been removed by new engineers.
James Dooley: Yes. Some of that legacy Google code has definitely been pulled out on certain things. That is where the fun begins.
Now, the big one. The latest Google leak. What is your view? A lot of people are shouting “See, we were right all along.” There is a lot of confirmation bias, people cherry-picking the bits that back up what they already believed. Is there anything genuinely juicy in there for you?
Kyle Roof: A few things stood out. Click data. There is mention of keyword density. These are things Google has repeatedly said are not part of the algorithm. In some cases they have said “and never have been”.
We now have a list of internal fields and attributes. Google has admitted this is their code. They claim they do not know exactly how it is used, but they have said the leak is genuine. How strong each factor is, or when it is in play, is another debate. But it certainly looks like they have been very dishonest about which inputs are possible.
I do not think you maintain a list of “things we do not use” inside the same codebase as things you do use. At a minimum, those fields existed and were plausible factors at some point. Maybe the generous reading is they used to use them and do not any more. But they have explicitly said some of these things have never been part of the algorithm. This leak shows that, at the very least, they were.
To me, it looks like these factors are in play at some times, for some queries, and in some SERPs. That means their public statements have gone beyond “a bit vague” and straight into direct lies.
A lot of us have felt that for years. Google would say “X does not matter”. You would test it. You would see the opposite. They were telling people to ignore things that did move the needle.
I saw some Google talking heads saying “I think ‘lie’ is a strong word.” I do not think it is strong enough. What is the point of having a mouthpiece for Google if the mandate is not to create clarity, but to provide cover and misdirection?
Their stated purpose is to help SEOs understand things. If they are happy to say things that are directly untrue, then their goal is not education. It is narrative control.
I understand they should not reveal the algorithm. I agree they are under no obligation to share everything. But once they start making absolute statements like “we do not use this” or “this is not a ranking factor” and that turns out to be false, that goes beyond PR spin.
I have always thought the analogy with Google Ads “helpers” is perfect. Those people are not there to help you crush Google Ads. There are only so many ad spots. They do not want to help you win all of them. What they want is more advertisers, all bidding at a mediocre level. That makes Google the most money.
It is the same here. The job of the public search spokespeople is not to help you do great SEO. It is to help you do safe, mediocre SEO, so nobody gets too much of an edge. So they are happy to provide misinformation if they think it balances the ecosystem.
I would call it lying. I would call it misinformation. I would call it disgusting. And I think the people still defending it know they are on the wrong side of this and are now trying to rewrite their own history.
James Dooley: Yes. It is crazy how it has all come out. Like you said, it is not the fact they keep some things back. It is the fact they straight up answered questions with “No, that is not a thing” when it clearly was. I do not mind them saying “We cannot disclose that.” I mind them pushing people away from things that actually matter.
Kyle Roof: I knew things were suspect when they started answering with “In my opinion”. You would see John Mueller say, “Well, in my opinion…” Why does he have an opinion? Why are we relying on someone’s opinion when he could just lean out the window and ask an engineer “Is this in the system or not?”
If he is giving opinions, it means either he does not know, or they will not allow him to give direct answers. In both cases it means he is not a reliable source of truth on how the system works.
James Dooley: It is easy to scapegoat John, but realistically he is probably being fed half-truths as well. I doubt he is sitting there inventing lies on his own. I suspect he gets partial information, or carefully framed guidance, from engineers and PR, then repeats it.
Kyle Roof: Possibly. But is it true that he is scrubbing tweets? I heard that. I did not check myself.
James Dooley: I have heard that from a few sources too. I have not sat there diffing his timeline, but if that is happening, it is not a good look.
James Dooley: Let’s move on. Artificial intelligence is everywhere now. Images, video, content. We are using it for lead segmentation as well. What are your thoughts on AI, specifically content first?
Kyle Roof: AI is a tool. The version we have now will not be the one we use in a year. So you cannot become tool-dependent. You need to be concept-dependent.
Now is the time to learn how to prompt it properly and how to use it. If you do not, you will end up reliant on some third-party tool that simply sits in front of the same models and charges you a margin.
What AI does best is process huge amounts of information quickly. That is why tools like POP, and other on-page tools, fit well with it. You could do all the manual counting yourself, but the tool speeds that up. AI is another layer of speed.
In my view, you should be using it as that first draft engine. AI content has basically killed the three-cent-a-word writer. The output is often similar quality to what you get from cheap content mills. For that tier, AI is a replacement.
But it still only gets you 70 percent of the way. You need to edit for brand, for facts, for proprietary data, for uniqueness, for tone.
AI does not know your brand well. It does not know your factual context well. Some tools now browse the web, but they still hallucinate. So I would use it to speed things up, not as a finished product. Anything that speeds you up is good.
And I would not lock into one model or one tool as “my AI forever”. This space will keep changing.
James Dooley: So you are definitely on the side that we should embrace AI, but treat it as a power tool, not a magic brain.
Kyle Roof: Exactly. But be careful where you use it. It is not a maths book or a science book. It is a language model. It can generate code and formulas, but I would be very careful relying on it as your source of truth in those areas.
Until we get a genuine maths or science engine, which these current models are not, you should not treat it as a calculator for serious work.
Repetitive tasks are perfect for AI. Things you do over and over. That is where you get an edge.
Right now I am playing with email workflows. I watched Gael Breton from Authority Hacker show how he uses AI to draft replies to routine emails. Stuff like “Got your message”, “Can you send a headshot and bio” and so on.
He has it set up so that when he wakes up, he has 30 drafts ready. He just clicks send. That is a massive time saver.
That is where I would focus. Repetitive work. Train AI to handle that for you so you can spend time on higher-value thinking.
James Dooley: Let’s get into debunking a few SEO myths. There is a lot of noise. People talk on stage or online with a level of certainty that does not always match reality. Is there anything you see repeatedly that you class as nonsense?
Kyle Roof: The red flag is always certainty without evidence.
If someone says “This is definitely how it works” and they show you nothing, no test, no case study, that is when you know it is fluff.
They do not have to run single-variable lab tests like I do. A decent multi-site case study is fine. “I did this 10 times and 8 reacted like this.” Brilliant. That starts a real conversation. I can show my tests. You can show yours.
What you see in a lot of big industry publications is “Is X a ranking factor?” followed by “Definitely not” because “Google said in a tweet that…” and “This feels like…” and that is it. No experiments. No data.
That is garbage. It might accidentally be true, but it is not trustworthy.
The myths I end up debunking most are “X is not a ranking factor” where the only “evidence” is one line from Google and vibes. We now know exactly how reliable Google statements are.
James Dooley: You have spoken before about “Ivory Tower SEOs” and the definition of spam. Can you explain that?
Kyle Roof: It took me a while to understand why some people saw me as “black hat”. I do not think of myself that way. I do white-hat work for real companies that cannot be anywhere near a compliance line.
I realised some people think if you do not blindly believe Google, you are automatically black hat. If you test something Google told you not to test, like cloaking, you are branded black hat. They do not distinguish between testing a behaviour and using it in production.
Then I watched talks where people were talking about spam and black hat as if they were the same thing. They showed a SERP, pointed at a result and said, “That is there because of black-hat tactics, so it is spam.”
That is wrong.
Spam and black hat are not the same thing. Spam is garbage content that provides no value. Black-hat is a description of tactics, not of the content itself.
If you had the cure for cancer, would you slowly build a site from scratch, or would you use every possible distribution method to get that information in front of people? You would use whatever you could. Some of those methods would be against Google’s guidelines, but the content would still be incredibly valuable and absolutely not spam.
If you spend $1,000 to get a parasite placement on Forbes, you are not going to upload trash. You are going to put your best possible content on that page.
People are saying “search results are worse” and maybe some spam has gone, but maybe some of those removed pages were actually very good information that just used aggressive tactics to get up there. What is left might be “compliant” but not better.
So you have these Ivory Tower people who only ever work on DR 70 sites and say “You do not need links, just publish.” That is true at DR 70, but ridiculous everywhere else.
They also equate “anything black hat” with “spam” and that is just not accurate. You can do something against guidelines and still produce content that is better than the “white-hat” stuff in that SERP.
On the flip side, they label you black hat if you simply do not treat Google’s guidelines as sacred truth. If you test, you are a heretic. That is silly.
So there are two big misunderstandings:
– Doing something black hat does not automatically make the content spam.
– Refusing to take Google’s word as gospel does not automatically make you black hat.
James Dooley: Let’s talk about the current state of the SEO industry. You and I share the same mentality on this. I always say more millionaires are made in recessions. You said on stage, “When there is blood on the streets, it is the best time to buy.”
I think we are in that phase now for SEO. There is a lot of negativity. Google leaks, affiliate sites hit, people panicking. Link building agencies across the board are down. A lot of them say revenue is down 35–40 percent because affiliates who were big spenders have been wiped out or have paused.
What is your view on where we are and what that “blood on the streets” line means in this context?
Kyle Roof: Around 2018–2019 you saw a huge surge in SEO. People were spinning up sites, starting agencies, launching services. Then Covid hit and everyone was stuck at home. They all started side projects. Most of those were SEO or content related.
It was inevitable there would be a crunch at some point. Helpful Content, AI, the March core update, they all just compressed the timing. It all hit in one big wave instead of gradually.
We are now in that recession period for SEO. A lot of affiliates have disappeared. They were heavy customers for tools, hosting, content, links. That spend has gone.
My view is there is more pain to come over the next three to six months. More people will exit. More small agencies will close. More site owners will give up and sell assets.
That is also where the opportunity is.
As people exit, they offload sites. Service providers drop out. When the field clears, those who are left have less competition and more available properties and clients.
If you can weather the storm, you will be in a great spot when things stabilise. Things always recover. They always have. When the upswing comes, being one of the survivors is incredibly valuable.
If, during the downturn, you can also pivot and serve new needs, even better. That is why we are offering more “done-with-you” and “done-for-you” on-page services tied to POP. Some people now do not have bandwidth to dig through reports. They need someone to interpret and implement, while they run their real-world business. That is a gap we can fill.
James Dooley: Yes. Even outside this current crunch, you see a repeating pattern. People hit a ceiling. They get to £10k a month and cannot get past it. They bounce between clients. They have no real scale.
I talk a lot about the three Ps: prioritisation, procrastination, and perfectionism. They stall people. What do you think causes most SEOs to fail or plateau?
Kyle Roof: A big one is pure inaction.
Worrying about SEO and doing SEO are very different. People sit in Search Console, watch rank trackers, refresh analytics, but never actually change anything.
When they do not know what to do, or there are too many possible moves, they choose nothing. Paralysis. Then they say “SEO does not work” when all they did was watch their charts.
On the scaling side, the ceiling almost always comes from a lack of SOPs and delegation. They are doing all the work themselves. If you are doing everything, you cannot prospect properly. You cannot take more clients without dropping balls.
You need standard operating procedures that a well-trained cocker spaniel could follow. Step by step. No interpretation. “If you see X, do Y. If you see Z, do W.”
The easiest way to build SOPS is to have the person doing the task document it the next time they do it. Then you review and tighten it.
The more you can push out of your own hands, the more capacity you have to sell, build relationships and grow.
James Dooley: I asked the community on Twitter and LinkedIn for questions. Got a load of DMs. Let’s do some quick-fire ones.
You can ask a Google engineer one question about how the algorithms work. What do you ask?
Kyle Roof: I want to know when concurrent factors kick in.
There are situations where you can push on one factor and nothing happens because other conditions are not met. I want to know: what is the first concurrent factor that has to be present before certain parts of the algorithm activate?
Is it a traffic threshold? A query class? A specific type of SERP? Some combination?
If we knew that, we could say “In this SERP, these factors are not in play, so do not waste time on them.” Or “In this SERP, everything is in play, so we need to push harder on these areas.”
James Dooley: That leads nicely to the next one. You previously said on a podcast with Gael Breton that “the absence of a ranking factor is the number one ranking factor”. Do you still stand by that?
Kyle Roof: Yes. That line actually comes from Lee Witcher, a very smart SEO.
The idea is this. Let’s say there are 10 relevant factors. Do you go all-in on 4 or 5 of them, or do you make sure you touch all 10?
Evidence suggests you are better off lightly satisfying all of them than over-satisfying some and completely ignoring others.
It matches tests I ran years ago. You cannot swap signals. If the top pages all have three H2s and three H3s, you cannot just do six H2s and zero H3s and expect the same result, even though the total number is six in both cases. You need to mirror the pattern, not just the volume.
So yes, fulfilling each factor at least minimally is more important than over-optimising a subset.
James Dooley: You can turn the dial up or down on one ranking factor in the algorithm. What do you change and why?
Kyle Roof: I would improve measurement of clicks on links.
If one link is on a DR 10 site and 50 people click it, and another is on a DR 80 site and no one has ever clicked it, which one is truly more valuable? It is the DR 10 link.
If Google took user clicks on links more seriously and weighted link value based on that, you would eliminate a lot of spam. You would level the playing field for smaller sites.
The original Google idea was that links are votes. A link means “I have looked at this page and vouch for it.” In practice, that is not how it works any more. It is money, outreach, networks, placements.
Real users voting with clicks is much closer to that original concept and would be a cleaner signal. The leak suggests they have the ability to do this, at least in some contexts. I would like to see that expanded.
James Dooley: Do you think the reason they do not fully do that yet is cost? Tracking and processing all that data?
Kyle Roof: Cost is definitely a factor. They probably feel they get an approximate outcome with the current system, for less compute cost.
They do have approximations like the “reasonable surfer” model, which estimates the likelihood a link is clicked based on position, prominence and so on. That is a prediction model, not an actual measurement. It is cheaper.
But yes, tracking real clicks in depth would cost more. I suspect that is the main reason they do not go all-in on it.
James Dooley: You mentioned reasonable surfer. Makes sense. But if you believe click data on links should matter more, do you also try to make sure you are optimising your guest posts to actually rank and get those clicks?
A lot of people ignore that part.
Kyle Roof: I think optimising guest posts is the right play. It is more expensive and takes more time, but you end up with links that actually see traffic and pass real value.
Most people just drop an article and never look at it again. You are better off treating those posts as real assets and trying to rank them. You then benefit from the traffic, the brand exposure and more meaningful link equity.
James Dooley: Content freshness. Is there a specific time frame where you say “update every three months, every six, every twelve”? Or do you just set up POP Watchdog and let that tell you when to move?
Kyle Roof: If you do not have any system, I would look at important pages every three months as a rule of thumb.
If you have PageOptimizer Pro, I would absolutely set up Watchdog and let it tell you when something meaningful has changed, like an intent shift or big ranking movement. That is better than blind schedules because it is reactive to the SERP reality, not the calendar.
James Dooley: Do no-follow links work as part of your link building?
Kyle Roof: Absolutely. But there is a concurrent factor. The pages sending the no-follow links need to be getting real traffic. If that condition is met, they can still be very useful.
James Dooley: Masterminds or conferences?
Kyle Roof: I will say conferences, for one reason. Masterminds drain a lot of energy because, to do them properly, you need to be fully engaged, helping people deeply, thinking on your feet. I am often also speaking or organising at events, so my battery is already low.
I do love masterminds when they are done right. But if I have to pick, I will take the conference environment, especially the hallway and networking.
James Dooley: Best SEO conference in the world, if you can only attend one?
Kyle Roof: It has to be Chiang Mai SEO.
I will plug SEO Estonia as well. If you cannot get all the way to Thailand, Estonia is easier for Europe and the US. But Chiang Mai is special. Eight hundred people in the room, plus probably another 1,500–2,000 SEOs in town just for the week.
You have got 60+ side events. Everywhere you go you bump into SEOs. It is electric.
James Dooley: Right. Let’s wrap it up.
We covered a lot. POP optimisation and features. E-E-A-T checks. Watchdog for content freshness. The SEO board game coming out. IMG training and tests. The Google leak. AI. Current industry pain and opportunity.
Kyle, where can people find you if they do not already know?
Kyle Roof: PageOptimizer Pro, IMG Courses, and kyleroof.com are the main places. You can get me through any of those.
James Dooley: Brilliant. I appreciate your time, Kyle. It has been class. I am sure I will see you again at another SEO conference very soon.