Barry and Steve's SEO and Beers Podcast is LIVE!

It won't end well with these two.

I'm pleased to announce that I've recently teamed up with Barry Adams to discuss all the latest digital publishing trends - whilst drinking beer.

Barry has been a good friend of mine for a number of years and we share the same passion for surfacing good journalism to our algorithmic masters. There's been a lot of confusion about LLMs recently and the role of SEO so we felt compelled to clear things up.

Along with publisher predictions for 2026, we discuss if AI Mode will become the default search view in Google, AI prompt tracking tools, Zero Click search, how you can influence without authority as an SEO - and a drunken night in RedRuth, Cornwall.

We haven't got a pre-planned schedule for our new Oscar contender, we'll team up - a bit like The Avengers (although who's who in that I will never know) - when there's been a significant number of updates in the industry.

We haven't even established a proper name yet - so this is where you come in. You can send us your suggestions along with topics you want us to talk about right here: https://docs.google.com/forms/d/e/1FAIpQLScmoVivBiFBVCaW4gSTPuoNhaUccypwFf6luo7Qj2C8ofNFeg/viewform

It won't end well with these two.

Here's a transcript of the show:

Steve

Hi everyone, I'm Steve and we want to welcome you to our new podcast, which we agreed actually over a beer, didn't we Barry? So really, this is very much meant to be like an informal chat about some of the biggest trends we see in SEO and publishing but with a difference because obviously we're to be drinking beer. So what could possibly go wrong? We'll see what happens!

One thing is that we haven't given this chat or this podcast a name yet because we want you to decide it for us. Not because we couldn't be asked to come up with a name, but we thought it'd be a nice thing to do, give you guys a chance to input. And if you've got any questions that you want us to answer on future shows, let us know. And we'll put all the contact details in the show notes below.

But we'll kick off with some introductions. Now, I  know that you've probably done this a thousand times, Barry, but if you just tell us about what you do, I'll tell everyone what I do and then we'll just chat about a few things people don't know about us.

Barry(01:07)

I'm an SEO consultant specializing in news and publishing. My funny accent is from the Netherlands, despite my very English sounding name. I am actually a Dutchman, but I've been living in Northern Ireland for the last 17 years or so, because my wife is from Northern Ireland. I'm a freelance consultant and you can hire me if you want.

Steve (01:31)

Cool, and please do. So yeah, so I'm Steve Wilson-Beales. So I'm an SEO and content strategy consultant. I've spent most of my career working with editorial teams, either in broadcast or the likes of Microsoft and music labels and, basically, advising just like Barry, on publishing trends.

So, what I was curious about Barry is how you got into SEO in the first place. I don't know if you've discussed that before, but I haven't heard of that.

Barry(07:06)

Yeah, I've occasionally been asked about it. It's been a bit of a long story, because back in those two years at uni that I did do, there was my introduction to the internet, we're talking late 90s here- and I decided to try and build my own website and I realized that building it and publishing it on the web doesn't mean anybody's going to visit it. You need to be findable, discoverable. So I sort of dabbled a bit in SEO before we even knew what it was called, just trying to do the basics to get listed in the Yahoo directory and on Demos and Altavista. You know, this is pre-Google days. And then of course when Google arrived I did link exchanges and keyword stuffing and all of that sort of stuff. But I wasn't doing that in a professional capacity, because in the first few years, probably the first decade of my career, I was a pure IT guy, managing server farms, doing first and second line support, that sort of stuff. 

But at some stage I thought this web thing is quite interesting, and I was working through an outsourcing company at the time that put me in jobs at Philips here in Eindhoven, the electronics company in Eindhoven where I grew up in the Netherlands. And I was basically putting charge of server farms that managed their corporate intranet. And because I knew HTML they also asked me, can you also update the actual intranet itself? Because I knew how to code HTML pages and was a little bit familiar with Front Page, which is what we used back then to build and update those pages. So I did that and later when I went to do different jobs I got a job as an international webmaster at an industrial company- and that's still the coolest job title I've ever had, International Webmaster. You know, this is back in those days when you basically had one person in charge of all aspects of the website. And that was part of my remit there and I thought, because I'd done it for myself for so many years, that I knew all about it. It was all simple, you just stuff a few keywords in and off you go. But then my company sent me to a conference in New York, Search Engine Strategies, in 2007 and I walked in there like I knew everything about SEO and I think within about 30 minutes I realized, wait, I know nothing, I am Jon Snow. This is like an entire new world that opened for me. Really, those three days at that conference in New York for me basically decided the rest of my career. Like, okay, this is awesome, this has so much more depth and intricacy and I want to learn this. I want to become the best SEO that I could possibly be.

Steve (09:26)

Yeah likewise. For me, I had a similar light bulb moment. I was working at an agency, ⁓ Sapient Nitro, I came to advise on content for a private healthcare company. They had one SEO in the building and they said they would do some analysis and give me some insights. And I sat next to them and they got out Screaming Frog, which I hadn't seen before. And they crawled the website, bunged the data into a pivot table and told me specifically the kind of content that would perform well or the kind of content that they weren't doing. And for somebody who would have done their competitive analysis by just going to different websites and noting down what I thought from an editorial point of view, this seemed to be like really amazing insight. That did it for me. 

So I've always been curious since then, and I'm sure you'll agree SEO sort of attracts people that like problems to solve..I then managed large teams in the broadcast area and when Facebook pivoted away from news publishers in 2017 I introduced a renewed focus on SEO into the newsrooms again. So, yeah, I've been hooked ever since.

Barry(11:46)

It's addictive, This SEO.

For me, it's this perfect combination of doing something which requires you o understand how things work and try to take things apart as it were, like reverse engineering the algorithms, which the computer nerd in me really enjoys, while still doing something that you know has a human front that people will engage with that. They find your website, they click through to your web pages and they read that content. So when I was working in IT, I always felt I was just doing IT for the sake of IT. I never felt that I was having a broader impact on the world in any meaningful way. And with SEO I find that for me that's that balance between the IT, the computer stuff, which have always come naturally to me, while still feeling like I'm making a bit of an impact. That people can engage and interact with what I'm producing, of what I'm being a part of.

Steve (12:38)

Yeah. Okay, so why don't we go into some questions that I've prepared in advance and we'll start with a nice easy one, Barry, for you, okay? Will AI kill SEO?

Barry(12:54)

No. Next question.

It's an interesting one. Partly the confusion that we have about AI and SEO is that a lot of people have latched on to this phrase called Google Zero, where they think Google's traffic decline will eventually reach absolute zero.

In one way I'm a fan of the Google Zero concept because it does show people that it's a zero sum game. You're going to have to work harder, but Google's traffic is not going to go to zero. It's just going to be harder to earn your place in there. I think too many companies have been riding on Google's own growth to grow themselves. Google's search volume has steadily grown for as long as Google has existed. And the amount of clicks that Google has sent to the web has also continued to grow as a result of that. And that has flatlined. Google isn't sending more clicks to the web. I don't think they're sending significantly fewer clicks to the web. I think it's more of a flatline now. But it does mean that if you had relied on Google's total volume to grow, fuel your own growth, you now see that growth slow down or even go down, you even decrease the traffic, because it's just become harder to earn that traffic. You have to fight harder for it. You have to be more competitive and work harder in the same space to just maintain your traffic. And a lot of people have latched on to AI as the critical moment where that change happened. But you know this as well as I do. Google's decline to publishers has been on the cards for the better part of seven, eight, nine, maybe even ten years. 

So it's not new. It's just that AI has accelerated that particular process. AI has hastened the decreasing of traffic that a lot of publishers get. And I think we need to be bit honest with ourselves here. I think too many publishers are now disillusioned with SEO for various reasons. AI is part of it, the algorithm updates is not a part of it, we have site reputation abuse penalties, all those things. Make it harder for websites to earn revenue from Google clicks alone.

So they're like, right, we need to do different things. But that doesn't mean they should just abandon SEO. I think that is a very high risk strategy. In the same way that, you know, Facebook definitely lost interest in news publishers and a lot of publishers left Facebook altogether. But a lot of them didn't and still managed to get decent traffic from Facebook because they managed to compete effectively. because a lot of other publishers left Facebook, they just managed to scoop up the competitors' traffic and the competitors' impressions. And I think we're getting to the same situation with Google. Here traffic as a total is flat, so you have to work harder to get your traffic there. 

And of course Google also has other areas like the Google Discover feed, which sort of pick up the slack to a certain extent, although that also becomes less and less attractive now because Google Discover is turning more and more into a social equivalent feed with all kinds of social media posts in it rather than just news articles which again means publishers are going to have a hard time standing out there and need to do different things to earn that traffic. But yeah, the hype bubble is definitely real and I think a lot of publishers are panicking and when you're panicked you don't make the right decisions. I think a lot of publishers just need to actually look at their data, take stock of where they are and make smart long-term decisions about where to spend their efforts. While at the same time allowing themselves the breathing room to short-term into slightly different strategies. I think you need to be a bit nimble and able to adapt quicker. And this is the irony of news, as you know as well, in such a fast moving industry. The organizations themselves are very slow moving.

Steve

Yes, think search suffered in a way or SEO suffered in a way because the search traffic used to come in and it would create like the bedrock of the monthly figures. But no one really interrogated it because it was just there. So, you know, a good 30, 40 percent of traffic from my experience.was just there with, as long as you cover the news, you then added context to it. You could manipulate the headlines or optimize the headlines to pretty much within minutes see a change in the rankings of that. So it seemed to be a template that one could, once you'd learnt with time how the real time algorithm worked or Top Stories worked. could almost predict the amount of traffic that you might get. 

But obviously, as you say, those days seem to have gone and we need to sort of like be aware of the new environment that we're in. And speaking of that, I think, you know, it's been a very busy time, particularly for SEOs, but pretty much all digital teams trying to get their head around AI and how LLMs work specifically. what's your, how would you kind of look at the differences between something like GEO or the many other acronyms that it has and classic SEO?

Barry(18:43)

I'll be entirely honest with you Steve, the moment GEO appeared as an acronym I smelt the snake oil running in the streets because we this was like people jumping on the LLM bandwagon before we even properly knew what ChatGPT was doing and how it was doing it and whether they had invented an entirely new discipline to sell to clients for a lot of money, know banking people's ignorance and uncertainty and so far that hasn't really changed that much.

Fundamentally, optimizing for visibility and citations in an LLM, which is this GEO label seems like an utter fool's errand. If your website depends on people visiting your web pages, optimizing for LLMs is probably the worst waste of resources you can possibly do. Because every single data point we have about how people use LLMs shows that they almost never click through to the websites that are being cited. In fact, they just stay in the ecosystem much more and do follow-up questions and all those sorts of things. So, with that understanding that LLMs do not generate visits to your website, why do you want to spend all these resources and spend a lot of money on expensive consultants to get citations in these LLMs that don't have any direct connection to revenue models. And yes, people say top of funnel, brand awareness, all of that, great! But there are other better mechanisms in place to get those things, to build brand awareness, become a top of funnel resource for your audience. And that's called classic marketing, which we know works, which you have better attribution models for.which we're very very experienced and proficient in. So why do we want to invest in this unknown discipline which, if I'm honest, the tactics that I see people associate with GEO is just spam, total utter spam, polluting the information ecosystem with spam content to basically manipulate the LLMs to use you as a cited source. Why do you want to do that for some ephemeral gain that you cannot quantify in any meaningful way?

It’s a very high risk environment that you're setting yourself up to fail in the long term because these LLMs will start fighting against the spammers and they might see you as one of the bad actors in that space if you've capitalized on these tactics and penalize you to death and just destroy your citations altogether. At this particular moment in time I genuinely, especially for news publishers, do not see any reason why you wouldn't invest

Steve (21:44)

What would you say about the objection that the referrals might be non-existent at the moment, but surely they will increase in time?

Barry(22:01)

Okay, show me the data. Are referrals from LLMs growing faster than they are growing in users? No, the opposite is actually happening. Traffic from LLMs is decreasing. The latest versions of the LLMs are actually geared to keep people more into the ecosystem. LLMs have no benefit to send clicks away from their platform. In fact, it actively hurts them. They want to just pump up their own numbers. There is no business model for LLMs to send traffic to the web. There is no model at all. That's not what their purpose is.

Contrary to Google as a search engine, which purpose was to make information accessible and findable, they have a slightly different model there and Google benefits from a thriving open web. No matter how much we criticize them, I do genuinely think Google doesn't want to destroy the web. LLMs really don't care. They just want to make sure that investors keep pumping money into their massively inflated stock prices. Literally, I would really love to see some data points that tell me that investing in an LLM visibility at this point is a smart investment. Also, I have to say this Steve, a lot of people say, yeah, it's good to be an early adopter. And that's just not proven true by any historical data point. In fact, early adopters almost always tend to lose out because they make the mistakes that the late adopters don't then have to make. 

It's usually nearly always better when a new technology wave comes along to wait and see and get a bit of a sense of what the landscape is going to look like and what actually works and what can drive meaningful growth and then jump in as a late adopter. All the big technology companies we're talking about right now, none of them were early adopters, none of them were the first to come up with something. They were all late adopters of the technologies that they built their success on. So we don't necessarily need to jump into the LLM wheelhouse right away, just wait until the bubble bursts, wait until the hype cycle is over, and see what the new normal is going to look like, because there will be a different ecosystem at the end of this hype cycle, and then decide what makes for smart investments and what doesn't. We're nowhere near that point yet.

Steve (24:02)

I think it’s an extremely confusing time because we’re still learning as an industry how LLMs work - but also because we've been flooded with lots of AI tracking services. AndI because a lot of people don't understand how LLMs work and they are looking at AI tracking tools to be the solution. One thing, I thought was quite interesting, Kevin Indig in his predictions for next year, for 2026, he was suggesting that there will be a lot of consolidation of the AI tracking tool in 2026 because their initial funding will run out. So it'd be interesting to see how that develops there, but obviously we have lots of those tools in market at the moment and they're creating a lot of conversations, which I think is probably the reason why AI or LLM optimization has taken up a lot of the conversation, whereas I think there needs to be more publisher conversations around diversification of what platforms they are on, a review of their daily output, and an understanding of what is their actual USP. All those traditional editorial things. So yeah, I think we have a little bit of a shiny new tool syndrome at the moment.

Barry(25:41)

Absolutely. I should also say that LLM tracking, prompt tracking, which is what these tools proclaim to do, is again a bit stupid because the LLMs are probabilistic word predictors. You do the same prompt five times in a row, you get five different sets of citations and only one of those will be used by a tracking tool and then shown to you, you're cited in this and then you do the prompt yourself you get an entire different set of recommendations. So it's not even that it's bad data, it's deceptive data, it's the wrong data. And if you're starting to make decisions on bad, wrong, deceptive data you're in for a world of pain.

Steve (26:28)

Yeah, let's park that for now, because we could go on forever on this topic. I think it's going to be an interesting next quarter, as I said, as some of the tracking tools consolidate in that area. So moving on to audience growth tactics that we've seen for various publishers. I thought it might be interesting to look at experiments or case studies that you've worked on that have either been successful or bombed.. So, I'll kick off with the value of having case studies to support your SEO strategy. 

 So what I found is that when you're an SEO and you're entering a new business, working for a new agency or in-house meeting new teams, you need to figure out how you are going to prove your credibility. And I've always found that starting small with a case study, I mean, it could be as small as a section of a web or a particular content strand, no matter what it is, it's to choose something that you can action directly or in close conjunction with a developer that maybe you've got on side. And then to report that back to the business. It's to show that you've applied a theory and it's worked and you have the evidence and the data rather than it being your opinion. So that's a small thing on the value of case studies. I'm sure you might agree there Barry.

Barry(28:19)

Absolutely, absolutely. think especially if you're new in an organization as an employee or even as a new consultant, if you can make some few quick wins that don't require a lot of investment on the side of the website or some like you said stuff you can even do yourself, you're sort of proving your mettle straight away. You're sort of showing the rest of the organization that yeah, okay, if we listen to this persont here will be some good things that come at the end of it. It allows you to build up credibility and because if there's some low-hanging fruit you can tackle this way quite easily, you're going to build a bit of internal momentum and build credibility so people listen to what you have to say so that later down the line when you want to do the bigger things that you no need to get done, you have some credit in the bank and it's easier to draw those things across the line. If you start straight away by trying to tackle the big things, you're probably going to hit wall after wall, because you just don't have the credibility in the organization yet, and you run into all kinds of internal political barriers that prevent you from getting stuff done.

Steve (29:16)

Yeah. Or tickets, right? What might happen is that your big project gets carved up into multiple tickets that then get stuck in the production schedule and you might end up with only getting half of it delivered, which may not even have any influence on what it is that you're trying to do.

One other thing is dealing with the unreliable  volumes of traffic that are generated by Google Google Discover for publishers, particularly at the moment. One thing I've noticed in the past when working with news teams is that the traffic that you get from Google Google Discover can become extremely addictive for newsrooms. And I've noticed that when the newsroom is more interested in Google Google Discover traffic, the other side of the organic visibility or Top Stories optimization kind of falls by the wayside. As we all know, Google Google Discover traffic is highly unreliable. So you get hooked on the spikes and then eventually that traffic falls away but then publishers start creeping into the negative because they haven’t been optimising for Top Stories because their attention has been diverted elsewhere. I'm not sure if you've seen that.

Barry(30:36)

Yeah,absolutely. Like I said, it's very addictive and it encourages bad habits. Google Discover lends itself well, unfortunately, to fairly clickbaity headlines and low quality journalism. And once you get a publisher on board with that, then they're just going to chase after the clicks and do whatever they feel they need to. And that's not proper journalism. That's the exact opposite of what Google wants to show in Top Stories.

I don't know if you followed the Google Webmaster Central Conference in Zurich, by any chance, but there were few interesting slides there that showed that while Google Discover is based on the principles of Google Search, it is only minimally dependent on the constraints that Google builds around its search index, which gives Google Discover more leeway to surface content from a wider range of websites and not necessarily only from high authority websites. So that's the idea behind Google Discover, that it gives traffic potential to many more different websites. But it also means that it's very prone to spam, because the authority signals don't weigh as heavily in Google Discover as they do in classic search and in Top Stories, which means you get all kinds of spam websites popping up all the time, which we see all the time, and loads of examples of really bad spam pervading people's Google Discover feed. And you don't want to fall under that trap as a publisher, because of the high volatility of Google Discover. And I have worked with many publishers who've become so dependent on Google Discover over the years that when that Google Discover traffic disappears, and eventually it does tend to disappear, they don't know what to do. They're just utterly panicked. And like you said, they've lost the ability to optimize for Top Stories. There's no backup strategy there. And they sort of have to reinvent their SEO wheel. And I always tell my clients to see Google Discover as bonus traffic.

So don't count on it, but it's nice to have, it's something that you shouldn't really build a strategy around. But because of the pretty big potential of Google Discover as a traffic source, that message almost never lands and publishers really go all in on Google Discover. With the end result, usually, that they over-optimize for Google Discover, under-optimize for Search, and at some stage, because of the effect it has on their headlines and often even their content quality, you end up in a situation where they're building up negative quality signals that impact their discoverer visibility with an algorithm update and they lose a lot of the traffic, but also then impact their ability to rank in Top Stories and they have to then recline that hill all over again. And I understand that's a very hard balance internally in an organization to not chase after a potentially huge traffic source with all your might just because it might at some stage backfire.

But you do need people in leadership positions who are willing to hedge their bets a bit and willing to look at the long term and sort of tell people, wait a minute, are we doing the right thing for our long term survival here? Are we just chasing after the next quarterly target? And that short term thinking is really what is killing a lot of publishers because they're just going to chase after the next big thing and not really caring about building long term credibility, long term brand value and long term audience loyalty.

Steve (33:51)

The only other thing that I was going to add in here was, as we all know, YouTube is a massive growth area for publishers. As was reported in the Reuters digital news report - although news consumption is going up across social media channels, it's kind of plateauing on YouTube which indicates that actually there's more opportunity for newsrooms to optimize in that area. 

And the only thing that I was gonna mention that I've seen from the past that's worked well is something I learned when working with The Newsagents video team. When working on your video titles it's quite easy to get tied up on keyword heavy titles for your videos. Sometimes, just a killer title that would encourage an emotional response is the one to go for. For example, rather than creating a video title like ‘Donald Trump's 2025 Speech on Immigration at the so-and-so venue’ I mean, it's got the keywords in there, but it's not as effective as a title such as, ‘Has Donald Trump Finally Gone Crazy?’ Right? Loaded with a compelling thumbnail that's gonna make people click. So.It's all about understanding, as I spoke about at the NESS SEO and Editorial conference earlier this year, around understanding that value and having the title in sync with the thumbnail of the video. So, yes, it's interesting around keyword optimization for YouTube. Sometimes you've just got to go for that killer title.

Barry(36:11)

Yeah, this is an interesting area and I loved your talk at Ness, by the way. I thought it was one of the best talks we had at the conference and something I've heard as a feedback from many other attendees as well. I think YouTube is part of a switch to multimodal, which a lot of publishers are doing, some better than others. And it requires slightly different tactics, like you've just said and slightly different understanding of how people engage with that content. Where in Top Stories, for example, a very straightforward factual headline tends to work really well, in YouTube you need a slightly different approach. Like you said, a good thumbnail with a very strong captivating emotive headline is probably a better way to go. And that can then also translate to other video platforms like Instagram and TikTok as well.

Steve (37:02)

Right, okay, so moving on to SEO myths we're tired of. So this is our opportunity to kind of debunk some of the bad practice out there or the bad advice that seems to be circulating a lot on LinkedIn at the moment. So I picked some of these at random, but what about ‘Reddit is the answer to all your woes’. Do you think that we should be pivoting everything over to Reddit? There's a lot of conversation about it.

Barry(37:30)

No. Reddit is always an interesting one because it's one of those user-generated content platforms where pretty much all content comes from users themselves and therefore has a bit more inherent trust than corporate content, which I understand. But at the same time people are wrong a lot of the time and will put absolute nonsense on Reddit in comment threads.

And I think search engines and LLMs also realize this and probably build a curation mechanism in there as well. Because Reddit, if you have ever visited Reddit or any activity on there, and I'm quite an avid Reddit user, you realize it's an absolute cesspool of horrendous opinions, blatant misinformation and outright lies amidst the gems that do exist there. So you've got to tread very carefully. On top of that, Reddit as a platform is almost allergic to corporate influence, that if you go in there with a commercial intent, you're just going to get your account banned within five minutes. So it's very hard to build credibility on Reddit there as well. I think if you want to do Reddit well, the best thing you can actually do is do things as a company very well, so that your users who are also Reddit users might spread the word for you. Rather than try to directly engage on Reddit, you might just go in directly. And if you have really good experiences with a customer, you can maybe ask them, hey, if you're active on Reddit recommend us in the relevant subreddits or things like that. And often your active subreddits, like the ones that I'm proud of, are primarily focused on rugby and video games, like I said, you tend to see a lot of the same websites being quoted and cited there, because they do things really well, and they write content really really well. And on the other hand, websites that write nonsense content or clickbait content tend to be derided on Reddit and cast in a very negative light.

I think Reddit is powerful because it's used as a source in LLMs to a very high extent, but you need to have a clear vision on how you want to engage with that in a way that doesn't come across as trying to astroturf the public perception there, because that has a real big chance of spectacularly backfiring.

Steve (39:34)

Yeah.I also worked in that area as well. It's a really tricky place. And even if you have direct contact with account managers at Reddit HQ, they can't give you the visibility that you might be wanting as a publisher because they don't control the subreddits, as you well know.

So it's really, it's about taking a step back and working out how you're going to approach this. What is the value you're gonna offer the Reddit community? Because what they're not going to want is for you to jump into a conversation and just copy and paste a URL for your news article. That's quite likely a good way of actually getting banned from a subreddit in time.I think you get a couple of warnings, but yeah, it's just not worth it.

Barry(40:41)

I do have an interesting tactic that I have actually seen work, which is if you're a fairly respectable publication in your niche, it sometimes helps if you go on Reddit to your relevant subreddit and take a few user comments and integrate them into an article. And that tends to increase the chances that that article then gets shared on that particular subreddit because people recognize it like, hey, they're talking about us. We are being read by so and so journalists from that and that publication. And that sort of helps you tap into that community.

Just by using their content in your own articles with, of course, proper citation of sources to sort of increase your popularity and put yourself in a good stead with those Reddit users.

Steve (41:15)

Okay, so what about another one? ‘ Keyword research doesn't matter anymore.’

Barry(41:30)

That's been doing the rounds  for the better part of 15 years, because, you know, the moment Google announces a knowledge graph, everybody's like, it's not about keywords, it's about entities. And yeah, true, it's about entities. But people don't type an entity into a search bar, they type a keyword into a search bar. So we're still relying on keywords, we still need to understand how people find stuff on the web. And guess what? That's by doing

keyword research. That's the only way you can find out what people are typing into Google and even in other search engines. YouTube is a very powerful search engine. You need to do keyword research for YouTube as well. You need to do keyword research for everything. What do you think LLM prompt tracking is about, what you talked about earlier? A prompt is a keyword basically. So yeah, you still need to do the research to understand what users are searching for, to use the right terminology, if you're using the right terminology on your own website, and if you have content that matches the intent behind those searches. 

And yeah, we can all talk about entity optimization as much as want and it's a very important part of doing SEO properly and also to do GEO properly. But it's still focused around words that people type into a system and we need to understand what those words are and what the meaning behind those words are for people who use them so that we make sure we match our answers, our content to the questions that people are asking.

Steve (42:48)

Perfect. I do want to move on to predictions for next year, but I thought we'll talk about publishers that we're most proud of at the moment. There's still a lot to be done in the audience growth area, but I'll talk about three.

I thought Zetland and Denmark are doing particularly great work around rethinking their KPIs. So there's a really good case study there where they have prioritized subscribers over other KPIs. If you go and read their sister publications, it's a really interesting case study in how to re-engage with their audiences. Newsrooms normally present quite a polished product to their audiences but at Zetland it’s quite different. They have a real emphasis on authentic photography. So a lot of the photography is taken from mobile phones or at least it's shot in a way so it looks like something like a mobile phone basically.

They are also constantly connecting with their audience to work out what should be the topic of their next article or the podcast output. So that's really, really interesting. The other two publisher that is doing some interesting things at the moment is Conde Nast who are doing a lot of work around EEAT and profiling their journalists. If you look at a typical article on Vogue, for instance, you'll see quite a lot of EEAT detail there, particularly if it's a review. There's a lot of explaining how the reviews are conducted, who are the editorial team, what is the editorial mission of that particular brand. And these are all signals, obviously, that Google wants to see, but also audiences want to see as they're trying to work out what are trusted news brands. 

And the other one I'll quickly mention is the Times. There are many other news organisations doing well, but I thought the Times conducted a lot of great audience research work. They’ve seen a decrease in search traffic and have gone: right, we need to start with the audience. Who is the audience? What is it that they want? And how can we align our brand with that? And not only deliver the content that they want, but also reinforce the fact that we are a trusted publisher.

Barry(46:12)

I want to highlight. One of them is a client, it's Media House Ireland, who last week launched an entirely new brand, Crime World, which they basically took the existing Sunday World news website, which was sort of languishing a little bit, and realized that a lot of the content that was very popular on Sunday World was related to crime news, both ⁓ small crime stories as well as bigger stories and investigative journalism around organized crime in Ireland. And they decided to take a bit of a gamble, with a new brand called Crime World, crimeworld.com, specifically focusing on crime stories in Ireland and the UK and the wider world. Which I thought was very bold, very brave, but I also think it's the right move to do. They went with the freemium model at the same time, where some of the content is free but most of it is behind the paywall, because they do feel they have something unique that they can offer to the Irish audiences that they want and that readers want and are willing to pay for. Combined with a really good website, very strong branding as well.

We're in the middle of that migration right now and I feel really positive about that move. I think it shows a level of courage from the publisher and the level of understanding of what the audience wants from them that not many publishers have. So I'm really excited about that project. 

And another one is The Athletic. I think The Athletic are doing really interesting things. I saw a piece of content that they published not too long ago about the FIFA World Cup draw where we're drawing the pools of the football teams and they had this really great interactive page built around that with predicting the games, predicting the outcomes, predicting the road to the final basically based on those pool matches. Which I think did really really well, got a lot of traffic for them and again that's one of those things. If you're quick as a publisher to latch onto things like that, if you have the skills in-house to build something like that and roll it out quickly and update it quickly, it reinforces, especially to the Atlantic (which is a hard pay world website) that people made the right choice to subscribe to them and made the right choice to give you the money because you immediately come with that very valuable useful content for people who are interested inthat really enforces that you understand what they want and that you can deliver that extra level of value beyond just listing the names of the countries in the pools.

Stevet's really, really interesting, the Athletic. I'm a recent football fan because as a kid, I always played with two left feet and my dad and brother were so passionately into football I was like, ‘this is not for me’. But my kids are hardcore Man United and Spurs fans. So I've come into football at a later stage in life.And what I like about the Athletic is that they catered for people who want to know more about Man United, teams, the challenges of the Premier League but it’s all highly accessible.

Barry(49:57)

Their newsletters are very very good as well. The quality of writing just draws you in. When you start reading an article, you start reading a newsletter, you don't stop until the end because it just draws you in. They get the tone of voice right, they get the level of detail right as well. They don't write content for beginner audiences, they write content for hardcore fans. And if you're a hardcore fan and you start reading that stuff, you're like, yeah, this is it, this is what I want to know, tell me more. They're really really good at stuff like that.

Steve (50:08)

Okay, let’s crack on. So, things to look out for for 2026. Now we've already talked about ⁓ platform diversification. I think it's a no-brainer that we're gonna see publishers invest more in things like Reddit, Social media, newsletters etc

Barry(50:50)

None of them individually I think are going to save a publisher - But you do need to have a diverse source of traffic. Because if one of them starts to decrease or drops out, you have others to fall back on. I would also add to that that diverse traffic sources alone aren't enough, you need a very healthy foundation of brand traffic, which is either people typing your brand name directly into Google or going directly to your website. And that for me is the absolute crux of the matter. If you have a loyal brand audience it doesn't really matter what the technology giants are up to, what the disruption in search and LLMs is going to be like, because you'll always have that baseline of trusted readers who come to your website. But that also means that as a publisher you need to have something that people are actively willing to seek out and engage with. And I've said this many times before and I'll probably say it another dozen times before this year is over.

If you are a forgettable commodity publisher that just writes the same stuff everybody else is writing, you're probably going to disappear in the next two years. There's no foundation for you to build your growth on. You need to have something that is unique and different, and moreover you need to make it very abundantly clear to your audience that that is what makes you unique and different. You have to put an emphasis on those signals that make you stand out as a publisher, that make you different and make you worth engaging with, then you're just going to be forgettable and forgettable publishers almost always end up on the losing end.

Steve (53:04)

Yeah, and  I’ll just talk about  an example, which I ran in the past which I think is relevant because I know a lot of publishers that might be listening to this might want to innovate but they are being restrained by specific targets. I ran a content experiment many years ago with the digital team at Global called PopBuzz. The reason it got signed off is because it was shaped in the form of a startup or a content experiment. It was agreed that it would have a limited time to prove itself, or in this case about three months, and it had a specific target to hit. And having the specific time periods and a target got it over the line for approval. So I think it's about how you put your business case for experimentation, but there's never been a better time to experiment right?  It's a standard practice in any editorial team. If the targets are not being hit at the end of each month or there seems to be a seismic shift in referral traffic - that is an excellent time to experiment because the content team needs to generate new formats, new approaches, new angles, to see if any of it sticks.

Barry(54:25)

Absolutely. You need to do things, if you don't do things differently, you're going to get the same result. So if you want to change the course of your organization, you need to try different things and experiment a little bit.

Steve (55:01)

Do you think AI overviews are a bigger threat than Google core updates? 

Barry(55:12)

AI overviews definitely have an impact on traffic for publishers. If there's an AI overview at the top of a search result, the amount of clicks just significantly drops, and we know that, and it's unavoidable. But at the same time, they're here to stay. They're not going to disappear, and we just have to work around them.

The beauty of AI overviews is that if there's a Top Stories box with relatively recent news, you almost never see an AI overview. And I think that's a deliberate decision that Google made to make sure that the latest, most accurate news is being shown there and the AI overview doesn't provide false information about the news event. Which means that as a news publisher you still have plenty of scope to optimize for Top Stories without suffering the impact of AI overviews.

It's just if your focus as a news publisher isn't purely on hard news, but you have a broader range of topics that you cover, some evergreen content, entertainment, celebrity news, things like that, they're much more prone to being cannibalized by AI overviews. Having said that, and I've written about this before more than a year ago, a year and half ago now actually, for a press gazette, when AI overviews were first announced and everybody was panicking, I said actually yeah, but they're not the biggest threat to your Google traffic.

Google's algorithm updates generally have a much bigger impact than an AI overview has. We've seen publisher websites on the receiving end of core algorithm updates, they lose 20, 30, 40 percent of their traffic. Then the next update, they lose the same amount all over again. And that's because they're not sending the right quality signals to Google or Google doesn't perceive them as being an authoritative enough website to deserve that slot at the top of a Top Stories box. So I think they're much riskier, they're much more of a threat.

And as a publisher you need to make sure you become immune to the algorithm updates, is generally done, not all cases, but generally done by becoming a very trusted, high authority, respected news publisher on the topics that you regularly cover. And not going down that journalism clickbait Google Discover, which we talked about earlier.

Steve (57:12)

Right, a big one. Do you think AI mode is going to become the default search view or are we gonna get something different?

Barry(57:21)

No, I think AI mode as a default search experience is too risky for Google. It's very risky in terms of basically destroying the web that Google relies on. Google isn't ignorant to the fact that they rely on the free and open web to constantly supply them with new content and that they play a very important part in the value exchange ecosystem.

And AI mode would kill that because AI mode doesn't send traffic to websites. Almost nobody clicks from an AI mode answer to a site resource. On top of that, the monetization isn't really there for AI mode yet. And I think because of the inherent ⁓ probabilistic aspect of LLMs, the keyword targeting that we get in classic search just does not extend to AI mode in any meaningful way either. It would become a much fuzzier advertising ecosystem, much more uncertain with the much higher chance of inappropriate ads being shown alongside AI mode answers. And that's just not something that's easily solvable and it might not be solvable at all. And therefore I think Google doesn't really want to go down that route because it would actually actively turn off advertisers from spending on Google ads. And there's a whole risk on the regulatory side of things as well, where especially the EU will probably say, hey, hang on a minute, you've built this AI system entirely on the back of all the content you've scraped but now you're basically also withholding the traffic that the content then should earn. And that would hasten the EU's antitrust actions against Google, which Google is already on the receiving end of. And lastly, users generally don't want AI mode as the default experience. AI mode user adoption has been quite low. You have a lot of people who use AI mode once or twice and then go back to classic search and never again visit AI mode.

Yes, it has its small core of hardcore users, like almost all Google products have, but I think the data that Google is now seeing in terms of user interaction, it shows that AI mode just isn't the right answer for it. And I think Google has a better tool up its sleeve to integrate AI properly into search results, which they're already doing in the Search Labs test called Webguide. I don't know if you've seen Webguide results, but those sort of structured, compartmentalized search results with short summaries at the top of each individual block, still quoting blue links with the option to expand into more blue links per subsection. I think that's a very elegant search result, a really nice integration of AI summaries plus very visible and clickable source links to the origin websites. I think Webguide is much more likely to become the default search experience than AI.

Steve (1:00:05)

Right, so this is the big one. This has been the final question and it's probably the most significant to SEOs, but do you think calculating the ROI of SEO efforts in 2026 is going to be the key challenge for SEOs or audience growth teams in newsrooms?

Barry(1:00:27)

I think the ROI of SEO for most publishers has been abundantly clear already. I think the risk is that senior executives in their blind AI panic lose sight of the ROI that SEO is already bringing them. And I think it's up to us as SEOs to remind them where the traffic and revenue actually comes from. And that sort of a cold water moment hopefully would inform most publishers that hey, Google is still by far the largest driver of traffic and revenue and that ignoring your SEO efforts in favor of the latest AI hype is not a smart move.

Steve (1:01:02)

I have to agree. In a world where we’re  still trying to understand how LLMs work, SEOs are perfectly positioned to help because they're already working with algorithms. Trust in your SEOs, they're going to help you connect with your audience.