Webinar! Brand SEO: Mastering Brand Authority in Generative Search

Join us for a panel discussion exploring the growing importance of brand in an era increasingly dominated by AI-driven search.

With the rise of generative AI and ever-changing algorithms, SEO has entered uncharted territory. The game is changing fast, and digital marketers who rely on traditional tactics alone risk getting left behind.

In this webinar, our expert panel - including Daniel Foley-Carter, Chima Mmeje, and Dixon Jones - dive into questions at the forefront of every marketer’s mind, such as:

• How does a strong brand help stabilize your visibility during Google's algorithm volatility?
• What’s the real impact of generative AI on branded search, and how do you adapt?
• How should your KPIs and analytics evolve to reflect AI-driven changes?
• What practical steps can your brand take today to remain visible and authoritative?

Webinar recording

Subscribe to the Sitebulb YouTube channel! You can also sign up here to be the first to find out about upcoming webinars. 

Webinar transcript

Jojo Furnival:
Hello everyone. So we are back with another webinar today and we've only got two out of our three speakers so far, but I'm sure that Daniel will join us in a bit. And if he doesn't, oh, well, we're going to have a good time anyway. So I don't know about you, and Dixon was cutting it pretty fine. [inaudible 00:00:25].

Chima Mmeje:
Give him hell.

Dixon Jones:
I was trying to work out the internet worked. I'm sorry about that, Jojo.

Jojo Furnival:
I know. Goodness me, you just can't get the people. So I have seen a lot of talk lately about brand building and the growth of AI search. So understanding how to master brand authority and generative search has become a super hot topic, I think you'll agree. And I've got whoever turns up today to discuss it with, but I've got two out three awesome guests, Chima and Dixon. So I'll introduce you guys properly in a sec, but before I do, my name is Jojo. Some of you know me. I'm the Marketing Manager at Sitebulb. It was far too hot to wear my Sitebulb t-shirt today. So you just have to contend yourself with Sitebulb blue necklace and earrings.
So Patrick, who is our co-founder and CEO, he's in the chat. Feel free to say hi to him, and obviously our guests too. Tell us where you are joining from. If you aren't familiar with Sitebulb, I'm just going to very quickly give you the low down. Sitebulb is a website auditing and crawling tool. In fact, we're the only crawler on the market to offer desktop-based software and a web-based cloud product. So that means our tool can help every SEO, from solo freelancers all the way up to global enterprise brands. And Daniel's here. Woo-hoo.

Daniel Foley Carter:
Hi, everyone.

Jojo Furnival:
Sorry to just address you directly. Shame you. No, I'm just joking. So you can try our desktop crawler for free. It's loved for its user-friendly interface and in-depth reports. Just head to sitebulb.com/download for the free trial. Patrick should share that link in the chat. We've also got Sitebulb Cloud, which is the same tool but accessible via your web browser. It is dramatically more affordable than other enterprise scale crawlers out there. So if you want to find out more about that or book a demo, head to sitebulb.com/cloud. Patrick will put the link in the chat.
Okay, just a couple of bits of housekeeping then we crack on. We are recording the webinar. I'll be sending the recording out tomorrow. So don't worry if you can't make it for the whole thing. We will have some time at the end of the webinar for your questions. So please put them in the Q&A tab, which is just next to the chat. Don't put them in the chat box itself. It's much easier having it in the Q&A tab. You can also upvote other people's questions in there. Okay, housekeeping is done. Let's get to our guests.
So first up, last in, we've got Daniel Foley Carter, probably the first SEO I ever followed on LinkedIn, an SEO evangelist since 1999. Director of Assertive, founder of SEO Audits IO and SEO Stack. He's one of those SEOs that has done it so long, he's built his own tool to make the job easier. If you are not following him already, make sure you do.
Next up, Chima Mmeje. I hope I said that right. I'm so sorry.

 

Jojo Furnival:
She's Senior Content Marketing Manager at Moz and a super opinionated voice in the digital marketing community. She's an industry speaker, founder and advocate for the freelance community for developing countries and a highly respected content and SEO strategist.
Last but not least, Dixon Jones, who was UK's Search Personality of the Year way back in 2012.

Dixon Jones:
Yeah, yeah. I think I could give that one up now. It's a long time ago.

Jojo Furnival:
No, no. I dug that one out. I dug that one out just for this intro. Dixon is a long-time friend of Sitebulb, and like Daniel who's been doing SEO so long that he's moved into the SEO tool space and is CEO of both InLinks, and now Waikay, which stands for What AI Knows About You. So that's our lead in. Welcome all. Thank you so, so, so, so much for joining us today. Right, let's get into the questions.

Dixon Jones:
Thanks for having us, Jojo, by the way. And sorry I was late getting here, but-

Jojo Furnival:
You're fine. You're so fine. I was here like an hour early because I'm like that. So we're here to talk about brand and its importance in SEOs, but has anything really changed since Eric Schmidt said in 2008 that brands were the solution to the internet cesspool? Is brand building more important to SEOs now than it was in the era of traditional search? Daniel, do you want to kick us off?

Daniel Foley Carter:
Yeah. First of all, thank you for inviting me on. And so in terms of brand, I think to be honest, brand equity has definitely become something that's of more value, especially since obviously AI is now the new norm and the buzzword, and now everyone is flocking to their thoughts and things that you should be doing to be cited and to appear in LLM-based queries. But I think it really goes beyond that. I think that there's quite a few facets to consider. So the first one would be that a stronger brand naturally is going to mean that you have exposure in other areas away from relying on just Google. I think anyone that knows me has known for quite a long time. I have a very strong stance against Google, not in necessarily a negative way, but I've seen Google go from what it was, a kind, friendly brand that was clear and uncluttered to a corporate giant that now obviously is doing a lot of damage to the creative community. And I think the monopolism that they've ran has made it so that there's not been enough diversity in other forms of search.
So I think when we think about that, brand can semi-compensate for that in some way. If a brand is strong, it's more likely to appear obviously in other areas of the web. And I think brand strength obviously will improve the probability of being cited. So the stronger the brand, the more likely it is to have naturally things like stronger links, more link presence, more mentions in UGC, and ultimately, I think stronger brands are also going to have a better chance of compensating for traffic loss from organic and diversifying from other traffic sources.

Jojo Furnival:
Yeah, yeah. Okay. Do the rest of you guys kind of agree, generally agree with that? It is definitely becoming more important?

Chima Mmeje:
Yeah, for sure, for sure. 100% agree.

Jojo Furnival:
Okay. Well, we've got... Oh, go on.

Dixon Jones:
I'd like to put a counterpoint in if I may. I'm not disagreeing, although it's much more fun if we disagree.

Jojo Furnival:
You can. It's totally fine.

Dixon Jones:
And I don't feel I've got a problem If I disagree with Daniel because we'll go to the pub and drink it out because we live about-

Daniel Foley Carter:
Oh, yeah.

Dixon Jones:
... about 500 yards from each other. Maybe a bit more than that, but I think that of course, everything that has been said is true, but I think there's still a little bit of truth in the fact that once you're in the LLMs, you're in the LLMs and that isn't necessarily a fraction of size. You've got more probability of getting in there if you are bigger, for sure, but it's much more about the accuracy of your context, I think is going to increase your chances of getting seen in the right place at the right time. And I think you've got a question coming up about it later on, so I won't get too much into the Carrie Rose one, but I know that-

Jojo Furnival:
Well, you can. You totally can now. What is more important?

Dixon Jones:
So you said before that you're going to bring up Carrie Rose mentioned that Red Bull was not doing so well in a particular context. Funnily enough, I hadn't seen that, but I also said the same thing at a talk last week in that... And I hadn't got a physical example from Red Bull, but I said that one of the interesting things about just getting your brand bigger is that it used to be, Red Bull did a brilliant... I think we can all agree that Red Bull's marketing over the last 10 years has been phenomenal. They've jumped out of balloons in space. They've got soap boxes on television programmes and go-karts and [inaudible 00:09:19], but what it has done is its talks to its audiences rather than talked about the product itself. So now an LLM has the opportunity to hallucinate and misunderstand what it's about.
So I wouldn't suggest that they should have done anything different. I think that moving forward though, saying what you don't do for a brand is almost as important as saying what you do do if you want the LLMs to properly understand what you're doing. That's my counterpoint.

Daniel Foley Carter:
I'd love to chime in here as well. So I think what's really interesting, and I won't go too far into it, but I'm now on week eight. So I've actually been studying and learning LLMs. I've been learning about GPTs. I've been learning about transformers because I think one important thing is it's easy to know how a car goes along. It has an engine in it, has wheels, but if you truly start to understand the mechanics of the principles behind these things and how they work, then you may have a better fundamental understanding of what's really going on below the scenes. So if we take one thing, for example, for however long, we know that a big part of SEO has always relied on link building because ultimately, like the whole principle of trust, it's a voting mechanism, and it's also the thing that Google needed to keep incentivized because Google could only grow its index by what it could find crawling the web. So if you incentivize link building, you create more and more avenues for Googlebot to discover new content.
But I think when we start thinking about LLMs and the fact that they're trained and trained and trained, and then that training allows them to have a larger and larger database of sources, as well as places that they can go and retrieve information from, I think then the model begins to change. So we're talking about link building, but then when we're talking about appearing in LLMs, for example, one of the things that I was learning about was vectoring and it was learning about how words are converted into numbers and those numbers basically mean that... I'm really sorry. Someone's knocking on my door.

Jojo Furnival:
Oh.

Daniel Foley Carter:
It basically means that like with vectors, you have similarity between words, which LLM infer. And I think the way that things are going, like with brands, how does an LLM know a brand's authority? What cements that brand as more of a go-to brand? Is it the quality of the information? Is it citations? What are these things? And I think this is the kind of stuff at the minute that we're figuring out, and obviously I've yet to play with Waikay, but I think that goes some way to really knowing what AI understands about your brand.

Jojo Furnival:
Well, maybe we can just actually just address that directly here. And I've seen a bunch of studies myself. There's definitely this kind of research is coming out now, but what do we actually know about what it takes for a brand to appear in an AI search result, whether that be the different LLMs? Because I know they all prefer slightly different things, or an AI Overviews/AI Mode. What can you guys tell us about that?

Dixon Jones:
Chima, do you want to go?

Daniel Foley Carter:
I was going to say, I'll put mine really quickly. So first of all, there's two key things. So number one, there's the pre-trained information that LLM has. So if it's a long-established brand where that brand's been cited in loads of sources across the web, during that training, if there's sufficient information for the brand, it may recall that. Whereas if there's something about the brand or a brand search with a variant I like a brand and the product and that isn't part of the training data, then obviously, an LLM may use retrieval augmentation generation to get that information.
So I think that's it. And I think if we talk about Google AI Mode or AIOs or LLMs, they all are operating quite differently. So I think the way to have a presence or to understand how you could improve your visibility in those, I think each of those are going to have their own conditions like with AI Mode and the query fan-out technique and things like that.

Jojo Furnival:
Okay. Yeah, go Chima.

Chima Mmeje:
So what we know, at least if I remove AI Mode from that equation, we know they like... No, we know they like brands. We know they like recognisable brands. We know they like brands that people have a positive experience with because they can interpret that by how people talk about your brand online on places like Reddit and things like that. So they're able to get that information. And we also know that they like links. In fact, there's a Ziff David study that showed that they reference brands that have high brand authority and high domain authority. So all of this under it is still driven by links and mentions. So we know that they like those two things.
And we also know that they like recent contents that is written by people who are thought leaders, first person, UGC, all of all that stuff. But what I find interesting is that it feels in a way, it's the same metrics that we're using to measure traditional SEO that is also being used to measure what is showing up in AI search.

Jojo Furnival:
Yeah, I suppose that's trying to figure out the difference between... There are very subtle and some people say they're the same and some people say they're not the same.

Chima Mmeje:
Yeah, I would say the difference, in a ways that there's a bigger emphasis on how people talk about you in the same way about how you talk about yourself. So a lot of the content that shows up on Google is content that is written by brands for themselves. So in a way, you can actually... I don't want to use the word manipulates. I want to say control. You can control what Google says about you with what you put out on your websites. Brands like HubSpot, all the big brands that we use as examples for good content marketing, it's from their websites. Right? But a lot of the time, when I'm looking for marketing resources, [inaudible 00:16:26] things like [inaudible 00:16:27]. The kind of stuff I'm seeing is from community. I didn't even know HubSpot had a community, but that is the stuff that I see gets incited more in ChatGPT versus content from HubSpot itself.
It's the same with when I'm seeing all this other community-based content. I'm seeing content written by people who are reviewing tools. I'm seeing an aggregation of.... Will I say an aggregation of sentiments? That is what I'm looking for, an aggregation of sentiments from review tools about how people talk about a tool when it's looking for what to recommend. These are things that in a way, they show up for local SEO, but it is more reinforced when I'm looking for recommendations or when I'm looking at content that is being, will I say attributed from ChatGPT? And I say ChatGPT because that is the one I use primarily. So I can't speak to the other ones.

Jojo Furnival:
Yeah. Okay. So everybody at this point has mentioned how important authority is and brand. So how the hell are small brand marketers supposed to keep up, stand a chance in this landscape?

Chima Mmeje:
Yeah, so that's the thing. I think I made a post about this on LinkedIn yesterday because I was so pissed off.

Jojo Furnival:
Yes, you did.

Chima Mmeje:
I was pissed off. I was pissed off because there's a lot of thought leaders talking about how to optimise for AI Mode and I'm like, "What the fuck are you talking about how to optimise for AI Mode? It's a new experimental thing. What have you done to be releasing anything about optimising for it?" Because how many people have even tested it to start giving us anything about optimising? And it feels very generic. There's a danger to that kind of generic advice that makes everybody think that they can go and do this thing in this checklist and they're going to show up in AI Mode. I have read that patent. We have a post coming out on the Moz blog tomorrow, written by somebody around that patent. I'm sorry, but I really don't think that small websites should be too focused on showing up in AI Mode.
Number one, why are you trying to show up in a platform that is basically stealing your content? That is basically trying to keep people inside your ecosystem and not send traffic back to you? You're a small website. You need all the traffic you can get because you need deals more than visibility. So instead, I would say you would have better opportunity on platforms like... depending on what you sell, on platforms like TikTok, on maybe even Instagram. Social media would be a good place to start. Paid ads right now, I never thought I would be the person saying this, but I think you would have better ROI, starting with paid ads to get revenue using social media to build your audience. I want to say not during communities, but as a small business, it probably don't even have the budget for your community manager or any of those things. But you can also launch your community with social.
SEO is something you do because you want to have a good website. You want to have all of that in place. You want to have content that people see that helps them make a purchasing decision, but you should not be prioritising AI search. You should not even be prioritising SEO right now if you're a small business. You should be prioritising the channels that allow you to reach your audience right now, and then SEO becomes a longer-term play. [inaudible 00:19:55], honestly.

Daniel Foley Carter:
I think I'd jump in and say it's a bit of a perfect storm here because the point that you make about AI Mode and optimising it for it, I agree with that. I think it's a double-edged sword because I think ultimately, businesses that just don't want to put the effort in because let's face it, the chance of getting a click is significantly lower as we've seen with CTR dropping off a cliff. So what do these businesses do? They turn to paid. So not only does Google keep more people in its ecosystem, it will drive up revenue because people aren't getting the level of traffic. And obviously, I think this is one of the reasons why... It's like never ever bad-mouth the hand that feeds you. Google has been my platform that has given me the chance to earn my living, but I think the fact is that-

Dixon Jones:
And you bad-mouth them every day.

Daniel Foley Carter:
I do now. Yeah, I do now. I do now because for me, I've seen the damage it's done. I've had people come and book an audit after HCU, which was like the biggest bare-faced load of rubbish and lies. Google lied. They absolutely lied. They knew what they were doing.

Chima Mmeje:
Yep. Yep.

Daniel Foley Carter:
They knew what they were doing and they let that go. And all the people that did everything that Google banged on about with EAT, making your content people first, blah blah, blah. Even though AI Mode is just a rehash of everyone else's content, they're doing the very thing that they put in their spam guidelines not to do. And I think the thing is that this whole ecosystem now is rotten. The fact is that it's now really about, okay, what is damage limitation? So stronger brand, broader marketing effort. Obviously, that being PPC, which is obviously still another side of the evil coin, but in terms of compensating, it's, "Okay. Well, we get less traffic. So better to convert more of the traffic that you got." So CRO should be more of a priority as well, even for small businesses, just making sure that you are converting as much of the traffic as you get.
And I think while we are losing traffic, informational traffic is obviously going to be the first to get hit. I still think there is still opportunity in commercial transactional stuff, but I do see that once AI Mode is lightfully rolled out around the world and people adopt it more and more, then we'll start to see more decay. So it's a bit of a funny time at the minute.

Dixon Jones:
I think we have to as SEOs look at AI Mode differently to the way that we have been. And I think that the narrative that we've given ourselves, we may have been a little blinkered on how to make, I'm going to say money out of AI Mode. I think behind me, I can see there's a glowing computer behind me, which I bought about three, four weeks ago. And in order to buy it, I'm self-hosting LLMs on it. And so I'm doing what Daniel's doing. I'm going on an eight-week course and trying to learn all this stuff and I'm trying to host NAN on my own machine with my own versions of LLM. And I knew because I tried to buy a $200 computer, it didn't work. I knew I had to buy a big computer to do this properly. So I researched using, let's say Gemini Deep Research to analyse the right machine.
I then went and used another LLM to say, "What's the website? What's the product on Amazon that's closest in match to this spec that they've come out with in ChatGPT, from Gemini?" I then used another LLM to check the spec that the Amazon guy was trying to sell me and say, "Can you get a better spec for an LLM hosting machine than this one?" And so I was doing all these things to analyse everything. Then it was only after I'd spent $2,000... Actually, 2,000 pounds on a computer on Amazon that I realised that I bought from a brand I'd never heard of, a website I'd never been to, and the computer itself had no reviews because nobody had bought that specific computer. It works. It's fine by the way. It's great, it's great for the job, but the point was that all of the research had been done through AI Overviews, LLMs, ChatGPT or whatever. It had done that research for me. Yes, the content had been read, but not by me, by the LLMs.

Jojo Furnival:
Yep.

Dixon Jones:
And had been interpreted by the LLMs. So my point is that content now has to talk to the LLMs as much as it does to a human being, and the human being may never go to the website to get the sale. Doesn't mean to say that SEO or a version of SEO isn't important. What I'm saying is that we have to look at things differently. It's not about the click at all anymore.

Jojo Furnival:
Yeah, so that's really interesting because James Cadwallader, again, probably butchered his surname, was speaking at SEO Week about the rise of the agentic internet. So it is exactly what you described. So users aren't even going on the internet anymore. The AI agents are doing it on their behalf. So from this world of user experience that we have been living in, now potentially it becomes all about agent experience. And as we've kind of been finding out and discussing, there are certain things that make website content more easily accessible to agents. What's it to AI? What is the future of the internet going to be, do you reckon? What websites? What are they going to look like?

Dixon Jones:
I imagine that Chima is going to have to get very annoyed about the way she has to structure a story. I bet that's what's going to happen because they like chunks of simple-

Chima Mmeje:
No, no, no.

Dixon Jones:
No? No?

Chima Mmeje:
No.

Jojo Furnival:
Oh, my God.

Dixon Jones:
Yeah. I'm not saying you like it. Yeah.

Chima Mmeje:
Why are we making their job easier? That's my own. You see the thing with SEO, I think I said this in my article where I wrote about should we stop creating informational content? And it's that the moment we start to write for an LLM, we've lost the plot already. The moment we start to write for summarization, we start to write for content to be picked up by a machine, we've lost the plot because that is how we got to where we are now. AI Mode is a response to poor user experience with search. We have to understand that as the underlying base. We are still thinking like SEOs. We need to start thinking like the users. And for me, [inaudible 00:27:11]

Dixon Jones:
I so want to agree with you, Chima. I so want to agree with you, but I can't.

Chima Mmeje:
And I think it comes from a place of fear. Nobody's ready to make that shift yet. We're still thinking about optimization. Dixie's brain is literally an optimised brain.

Jojo Furnival:
Can I say, I'm loving this? I don't think we've ever had such a hectic debate on so many disagreements.

Dixon Jones:
You need it. This is the thing. This is new territory and if we're all going to come in with the same thing, we've got nothing. It's important to have different opinions, especially at this particular juncture, right? Because we're not going to make headway as a-

Chima Mmeje:
Yes [inaudible 00:27:53].

Dixon Jones:
... SEO community if we don't learn new rules, new paradigms to work with, and that requires debate. And if you're not scared, great. I've got three people that I know I can have a debate with here, which is great.

Daniel Foley Carter:
Will someone think of the link builders? No one here is thinking of the link builders. What are the link-

Dixon Jones:
I've got something for them. So I've got to scrap for them. So for me, what is interesting is that in all those LLM responses, because we're tracking it in Waikay, the citations then become very interesting. So we see what citations all the LLM uses, and I know Crystal Carter from Wix will back me up on this. Some of the citations are pants. It's like the listicles of 2010. Some of the citations that are coming out of the ChatGPTs of this world are poor, shall we say.

Chima Mmeje:
Because people are gaming it now. That's why.

Dixon Jones:
However, if you start doing that with your competitors and seeing what citations are coming back from your competitors, so you say, "What is the best computer for an LLM?" And a whole bunch of websites come back with the top 10 PC world computers for LLMs or whatever, and you've got all those. Now when your competitors start getting... You say, "What is the best Hewlett-Packard computer and your Dell?" Then you can go and see where they get mentioned and stuff, and then you can start going and getting yourself cited on those because it's not about the link anymore. But it sure as hell is about having your brand mentioned in the place where the LLM looks.

Chima Mmeje:
Yeah.

Daniel Foley Carter:
Are we going to see a new era of EMBs, exact match brands?

Dixon Jones:
I doubt it. I doubt it because an exact match brand, it becomes using words instead of... So Rand went on to SparkToro. He used a word that was completely made up. Waikay is completely made up. I've got words that are made up. They seem to be able to get hooked into LLMs better than Majestic or InLinks, which are words that can have other meanings. So I don't think so.

Chima Mmeje:
Yeah, I agree with Dixon on this because if AI Mode is looking at meaning and not the exact words, then exact match brands are not going to work. But also to your point on citations, I think I read a post from somebody last week or two weeks ago that people have started gaming mentions in a way that we get them cited on LLMs. Basically, the same stuff we're doing with SEO and black hats... Or they said we should not use black hat anymore. What's the inclusive term? I don't know. And that's hats from 10 years ago or whatever it was. That's part of what we're seeing now with everybody basically doing copycats. I don't have an answer to that, but it's a problem.

Jojo Furnival:
Okay, we've gone so off-piste, I'm not sure where the mountain is anymore, which is fine.

Daniel Foley Carter:
It's fine. We can head to the beach.

Jojo Furnival:
Yeah, we're all heading to the beach. That's fine. What I want to try and do is-

Dixon Jones:
There's loads of questions in the Q&A as well.

Jojo Furnival:
I know. Oh, bloody hell. All right. So I know that I want to try and leave the wonderful people that have come into this webinar with some practical tactics, potentially, concrete things that they can try and do. Maybe we're going to just potentially part... Well, I don't want to open up the debate about small businesses again, what they should be doing. Maybe just fuck off to social, I don't know. But let's just think about what new tactics marketers can adopt to increase that direct engagement with the customer if indeed the user, if indeed they can?

Chima Mmeje:
I can go first with this. So there was something that somebody talked about last night at our Sip and Search event, and she gave an example of how Nike, all their ads, if you look at Nike ads, is always that underdog that came to the top. That's Nike's story. It's baked into everything that they do. The ad with Serena Williams, the ad with... I've forgotten her name, but the athlete who does gymnastics, all their ads, if you notice it, that is baked into their story. I personally give an example of brightonSEO and how the brightonSEO story is from a room above a pub. Everything brightonSEO does, that text is baked in there. The Red Bull you gave is baked into my brain. Red Bull gives you wings. Every time I drink a Red Bull, when I still used to drink energy drink, I used to literally feel like I could fly because they have baked that into their story.
So if you are a brand and you are trying to get more attention, more visibility, start by identifying your brand story in a few words. If you think about the underlying thing that is your backstory, that is your underdog story, that if people hear that, it's going to make them feel affinity with you, identify that and use that in your messaging till thy kingdom come. That's the first step. Number two, start getting your internal Smiths outside. I give an example of Gong all the time. They do this so well. I discovered Gong through Udi Ledergor. I did not know what Gong was until I started seeing Udi's content on my feed. And then there was somebody else, somebody Reid, who also worked at Gong, and then they had a couple of more people.
HF is another good example. I know I work at Moz, but HF does a good job of bringing out their internal Smiths to be thought leaders on LinkedIn and that gives them a lot of visibility.

Jojo Furnival:
Right.

Chima Mmeje:
If you have internal Smiths, get them outside. Partner with communities as affiliates. Communities tend to be very, very engaged. Actually, communities like Women in Tech or communities that are on LinkedIn. Find the communities that are very, very active, and then build partnerships with them. Sometimes those partnerships don't even have to cost a penny if there is value and if it feels authentic, people will still engage with that. And then if you have the budget, get on affiliates. Affiliates are great for driving demand. And I say this because we have, in fact, because we've used affiliate and we are seeing the revenue from that. Affiliates will drive demand, affiliates will give you brand mention and affiliates will show trust in a way that allows people to see the outcomes, which is what you want for your product. So if I could say a few things that brands should be doing to get started with, that would be the trajectory to move forward in.

Jojo Furnival:
Okay. All right. Yeah, go Daniel.

Daniel Foley Carter:
Sorry, I was going to give my two cents on things that could be done or whatever [inaudible 00:35:12].

Jojo Furnival:
Yeah. Please do.

Daniel Foley Carter:
Okay. So I think for anyone that I speak to now in regards to SEO, I think panicking and looking around at all of the stuff going around on social media is probably the worst thing to do at the minute. We all know, as Chima said, you've got all these people that are out there peddling crap about how to dominate AIOs and dominate AI Mode and they've got no clue. They're just riding a bandwagon and using complex terminology. But the one thing that I would absolutely say is to go and set a couple of hours aside each evening like I've been doing, learn how GPTs and LLMs actually work.
If I could refer, there's something called 3Blue1Brown. So it's the number 3, Blue1Brown. He's got loads of YouTube videos and he explains everything from the ground up in terms of how GPTs work, how context work, how vectoring works. And when you watch the videos and then you learn the concepts, then it becomes a lot easier to understand what's happening when you are running a question search on ChatGPT or Claude or Perplexity. So the more you really understand the fundamentals, the more you can actually start thinking about how those fundamentals apply to your content. And one of the things that really did it for me was I was watching about how vectoring works. So go and watch 3Blue1Brown. He has a video on vectoring and it basically explains like words are given numbers and they're grouped together in how relevant they are.
And then it explains the example of context. So for example, if you had the word queen in an area with words around it, you would think like queen, king, bishop, pawn, or you could have Queen the band, and basically explain how LLMs have to work out context from one word to the next. And then once you start understanding really how these things are actually working and generating the information, the better you're going to have an idea of the kind of concepts of how you can actually make your content so that it's more likely to be cited. One thing that I think is really important here is that LLMs initially, like the early versions of GPT, GPT-3 onwards, they were trained on information off the web. And if you remember like in the early days of ChatGPT and you asked it a question, it would give you an answer, but it would be like, "That answer was out of date."

Jojo Furnival:
Yeah. For two years ago, it was like.

Daniel Foley Carter:
Yeah, then retrieval augmented RAG came out. I always get my tongue tied, where all of a sudden, it can go out onto the web, get fresh information back, and then obviously, it runs its process vectoring. Everything goes through like transformers, and then it can work out, "Okay, these things are most relevant." So fundamentally learning all of the mechanics underneath are going to make the onward journey for, "Okay, so what do we need to do? How do we maybe better structure our content?" So for example, creating shorter forms of passage content with context.
The other thing I would say as well is that brand popularity could be a thing. We know like ChatGPT leans on Bing, for example. Indirectly, Bing might lean on links the same way that Google does, obviously to a different degree. So links remain important, but links might not necessarily be the thing that's influencing citation in LLM. So my point is always learn, and if you see people peddling like, "Oh, this is how you rank and this is how you dominate AI Mode," when even experienced people in the industry are still learning, don't get lost in the noise.

Chima Mmeje:
Yep.

Jojo Furnival:
Yeah. That's fair. Dixon, go.

Dixon Jones:
I actually agree with everything Daniel said and I don't want to repeat any of it, but Waikay does a lot of that of turning content into entities so that we can then do that vector matching and things. And so one way that you can do that without using something like Waikay though is what you can do is to find content gaps that the LLMs need to get hold of to give you that edge is to say, let's say you are Dell or Nike and Reebok, then what you could do is you can go to... If you're Nike, you can go to an LLM and say, "Why would somebody buy a Reebok sneaker instead of a Nike one?" And that's an interesting prompt because it's sharing.
The LLM, if it's going to not hallucinate, is going to come out with the things that Nike is not doing. It's coming out with effectively the content gaps or the feature gaps. So that's the underlying principle of how we're approaching it. So I've got to put a link in the comments for those that are there to an action plan that does it for the concept of sneakers and how Nike might do some things differently, but it's all about finding out what's unique about your competitors and shouldn't be? The things that your competitors are getting noticed for in LLMs that you aren't is where you'll get your context.
And then the other part of that is making sure that you are crystal clear about what you do offer. If you're a barber in Lincolnshire, UK, then don't pretend that you can cut hair in New York and don't pretend you can cut mine because look what somebody's done to mine.

Jojo Furnival:
So that's really interesting because it's becoming more and more and more and more important, even if you don't prioritise, even if you don't put all of your eggs in the AI basket as it were, it's still really important to monitor how your brand is being cited and perceived and retrieved, et cetera, et cetera. So I imagine that there's probably already been numerous... I'm sure Lily Reyes had these situations, but what can brands do if they find that AI is misrepresenting their products or values?

Dixon Jones:
We're doing it all the time. So there's an agency on the West Coast, Los Angeles, Hennessey Digital, and he owns hennessey.com, with an EY.

Jojo Furnival:
Oh, yeah.

Dixon Jones:
And I put him through the system and it said he was two things. Firstly, it was a Cognac, which is Hennessy without the E. And then also, a auto parts company because there was another brand in Los Angeles, it's an auto parts company. And back in the day, Jason Hennessey went to that company and said, "Hey, give me a link from your website saying, 'Are you looking for the SEO company? Go here.'" Brilliant idea at the time. Now it just says SEO company on an auto parts website. So this is the kind of problem that does change as the world changes.
So if you've got generic words, that happens quite a lot. So fact-checking is job number one. Yeah, absolutely. You've got to check the facts, and then track it back to the errant page on the website that caused the problem. And it may be the LLM that caused the problem, but you should be able to try and get yourself back to the source and fix it.

Jojo Furnival:
That's really, really interesting. He wrote an article for us, actually.

Dixon Jones:
Who? Jason? He did give me permission to talk about this publicly, by the way, Hennessey, the guy.

Jojo Furnival:
Oh, did he?

Dixon Jones:
I'm not out here. He thinks it's brilliant as well. It's really interesting find. Oh, I had another one with InLinks. So InLinks apparently got bought by Semrush. That wasn't true either. Oh, no, no, no. Well, that wasn't the hallucination. It said it was a rank checking company and we don't rank check. And we pulled that back to a broadly positive review of InLinks, but it added this feature that we didn't do.

Jojo Furnival:
Okay.

Dixon Jones:
So in that case, we could reach out to them and they immediately fixed it and we haven't heard about it in the LLMs again. Things are not always that easy to fix, but another thing I found interesting, Microsoft blocks Gemini altogether. It doesn't stop the LLMs from saying, "What do you know about Microsoft?" Azure or whatever it may be. It's just that they're losing control of their narrative, in my opinion, if they do that.

Jojo Furnival:
The LLM is never going to turn around and say-

Dixon Jones:
"I don't know."

Jojo Furnival:
"Sorry."

Dixon Jones:
"Never heard of Microsoft."

Jojo Furnival:
"We're not sure. Can't tell you anything." Go on, Daniel.

Daniel Foley Carter:
Oh, sorry. I was on a different track there. I was going to say the other thing as well is that for every positive thing that happens in this world, something negative has to come along with it. And unfortunately, I think we're also going to see the opportunity for people to manipulate and influence LLMs negatively. There has to be breaks on things somewhere, but with the automations that exist out there, it's not hard now for people to distribute content that might be negative or misleading of the brand, ultimately like poisoning.

Chima Mmeje:
True.

Daniel Foley Carter:
Because we've seen it where we had years of negative FCO until Google finally got to the point where it would just ignore link spam. I think now they're going to have the challenges, the gatekeepers to all this data. And one other thing I think is also interesting, I don't want to chuck too much in here, but have we also thought about personalization? Because I think-

Chima Mmeje:
In what respects?

Daniel Foley Carter:
Memory. Because obviously-

Chima Mmeje:
Yes. Oh, my God. Oh, my God.

Jojo Furnival:
Oh, yes.

Chima Mmeje:
Yes.

Daniel Foley Carter:
Yes.

Jojo Furnival:
Okay. That's probably a whole other webinar.

Chima Mmeje:
Yeah, [inaudible 00:45:41] memory in AI Mode.

Daniel Foley Carter:
Yeah. Well, the thing is that, like ChatGPT now has stored memory. So even if I go on and start a new conversation-

Chima Mmeje:
Remembers. Yeah.

Daniel Foley Carter:
If I've asked it something that I've had some similar conversation with it in the past, it recalls that and it influences the outcome and I've already seen that happen. So that is also going to shape forward strategies in a way, like what-

Dixon Jones:
As we're building tools, I think that this is going to make us have to use... Well, so the API generally doesn't have that long-term, short-term, it's called LTSM, long-term short-term memory. And so you then get an unbiased pull of data using the API, or I think you do. But yeah, it is a problem. Of course, individuals create their own echo chamber in ChatGPT now.

Jojo Furnival:
So guys, this is great, but we've got so many questions, as you pointed out, from the attendees to go through. So if you don't mind, we might switch to that now, with my glamorous assistant, Patrick, could pick one. Here we are. So, "We've seen a tremendous decrease in clicks with AIO. My thought was to transition from writing from informational intent keywords to commercial keywords. Is this the right move?"

Chima Mmeje:
Okay. Yeah, sorry, go ahead. Go ahead, [inaudible 00:47:22].

Daniel Foley Carter:
Okay. So I hate to spoil the party here, but obviously, SEO Stack has over 100 million keywords growing now. And in March, on the March update, we saw AIO prominence significantly increase and it was literally spreading across commercial and navigational queries. So whereas we had the initial two-phase rollout of AIOs last year, the impact of them was progressive, but then on the core updates, they've significantly increased the prominence of AIOs across commercial and navigational, and even transactional queries. I've seen them now, and not only that, they're unstable. So they appear in some searches one day, and then they don't appear again the next.

Chima Mmeje:
Yep.

Daniel Foley Carter:
So it's almost like nothing is safe from it. I think what I would give in terms of advice is something called compensationary strategy. So what I mean by that is if it took one well put together article to generate 100 clicks, you're now going to need 10 well put together articles, just using this as an example. And obviously, each article requires significantly more effort now because of what's happening with Google's indexing policies. So whether or not you're deterred or put off, you should still make a concerted effort to keep creating that content even though you don't get the same level of reward.

Chima Mmeje:
Yep. I just dropped that link in the comment session that I have content that literally answers that question. And basically, yes, you should still be creating informational content because you don't want to just keep jumping strategies, tweaking strategies every time there's something new happening. Your SEO and content strategy should be resilient, should be future forward. And if you just keep jumping every time there's something new, you are never actually going to see any sustainable growth. So seconding what Daniel said, you should still be creating informational content. Read the article. It shares how you can do that and get hopefully some better results.

Jojo Furnival:
Awesome.

Dixon Jones:
I'll say follow the money. You write about what it is that's going to make you money.
If you're selling flights from New York to Paris, then that's what you talk about because Google will never actually be able to fly you from New York to Paris. Okay? That's the truth of it. You talk about the thing you can sell. You can talk around that subject for sure, but if you are just talking about information, yeah, I'm afraid the ship has sailed, but there's already a lot of books in the British library and Google's already scanned them all.

Jojo Furnival:
All right. So, "What metrics/tracking are necessary to get good overview of brand visibility now? I'm newer to brand visibility and I want to wait to communicate its value with supporting metrics to clients."

Chima Mmeje:
I would say for me, priority would be referral traffic because that is the direct indicator of what is coming from all that visibility outside of organic. So number one for me, referral traffic.

Dixon Jones:
So Waikay is doing it differently. So what we're doing is we've given up on the click all together. We don't care about clicks anymore or traffic anymore. For brand visibility, what we're thinking is... And we're launching this thing tomorrow, it was supposed to come out today. So it isn't, but anyway. So our brand visibility tracker will say, "You put in the prompts. So what is the best site for sneakers? What is the best site for..." Whatever your prompts are. It'll then track every single... Well, every day or every other day on the engines that you want, see what brands are mentioned and start to track the visibility of those brands. So what are the best sneakers? And Nike, Reebok, whatever it may come out with. So that's our solution, is track the prompts and then see the brands in the prompts. And we think that's going to work pretty well.

Daniel Foley Carter:
I'd say tracking brand visibility. The answer to that from my perspective would be are you tracking your own brand visibility or are you trying to gauge external brand visibility? So for internal brand visibility, I'd still be tracking impressions in GSC. So I'd be looking at the volume of search. I'd be looking at external stuff such as Google trends to look at brand demand. I'd also be looking at offshoots of brand mentions. So some of the things could be looking in link tools. Obviously you've got Moz, Ahrefs, where you can look at keyword data.
The other thing I also typically tend to do is I use the in-text command in Google. So I use in-text colon, then I put the brand name, then I use the tools in Google Search. So you can use in-text the brand name, and then you can set using the tools, the mentions in the last 24 hours, 48 hours. And the good thing there is that you can actually look at how many results are returned. And I actually did write a script using Puppeteer that would do this. Obviously, I'm not saying that this is something that everyone can do, but just using the in-text command and looking at frequency results is actually good to see if you're popping up in new places, like in UGC communities. So I would typically tend to say brand visibility is a bit more of a fragmented thing.
And then lastly, obviously, you've got LLM tracking. I think some people use something like Profound, which I've not used it personally. It looks like it does look quite expensive, but I'm not going to comment on that. But there are going to be ways that you can track brand visibility using LLM trackers. One thing I would say there, just before I finish up, because I always make answers far too long, is that... And I think Mark Williams could have pointed this out, is that tracking mentions in LLMs isn't always efficient because LLMs are so inconsistent. So you could be querying LLMs for a position that you get, but then if another 100 people search for the same thing, they've got personalization, they're going to get all different things. So it's a bit of a difficult one.

Jojo Furnival:
All right. "With there being many similarities in the tactics you'd use to optimise for SEO and LLM traffic, how do you position this to clients and businesses? How do you differentiate your packages and outputs? Is it prudent to measure both channels together?" That's a good question. Who wants to eat that one?

Daniel Foley Carter:
I'm happy to. I know I talk a bit too much. So in terms of like LLM, SEO, AI SEO, GO, I don't know how many acronyms are being thrown around now. I think some people were enjoying it, some people were just getting fed up with it. First of all, sorry, can I see the question again? Sorry, because it was like more than one question.

Jojo Furnival:
Patrick.

Daniel Foley Carter:
So in terms of tactics, I think they just naturally share some of the same tactics. Like LLMs can't at this point render JavaScript. So you need to make sure that your content output is readable to them for RAG, but how do you position this to clients and businesses? Well, I think one of the really important things here, and this is why the industry is being polluted at the minute because the new shiny object that now clients are like, "We want to be in AI search. What do we do?" Well, the SEO community, a lot of it thrived on client ignorance. Not to put the industry down, but there's been a huge volume of client ignorance or lack of knowledge. I don't want to put it in an unfair way, but if people don't understand something, they're easy to prey on. And obviously, there's people out there want to make money and they don't care about integrity or ethics.
The reality is that if you have a good running relationship with clients, you just explain to them how it is. We are learning about factors that influence LLM visibility, but right now, we can't say categorically that doing A, B, C is going to lead to that visibility. We will adhere to best practise. So we'll do all the things that we know from a logic point of view makes sense. We'll strengthen brand. We'll look at brand mentions, brand salience and sentiment, searches and things for the brand positive. You do all the things that are common sense that would benefit SEO and likely have some form of benefit to LLM visibility. And educate the client. Don't say to the client, "Here's an LLM SEO package." Make it look like something unique is being done when really, it isn't. And that would be the way I'd position it.

Jojo Furnival:
Have we got time for one more?

Dixon Jones:
I was trying to answer that one and saying, I think you should measure them differently. And one of the reasons I say that is that if you're an agency, it gives you a new product to sell. Great. That's something else. Don't sell they're same... Sell something new. But also. We're doing searches on our phones. This app has DALL-E, ChatGPT, DeepSeek, Gemini, loads of things all in one app. I'm using an app. I'm not using the phone. It's a third party tool. So I'm tracking it in a different way. So I think I'd still call it SEO, but in the same way there's SEO for Facebook and SEO for YouTube and SEO for whatever, there's SEO for LLMs, each one of those LLMs.
This is great. We've been wanting for such a long time Google not have a monopoly. Now we've got a splinted operation where everybody's searching different things on different places. We should be rejoicing, guys. We should be rejoicing it in the industry, instead of lamenting the loss of clicks.

Jojo Furnival:
Okay. Okay. I was going to say on that note. No, "The one thing a marketer must..." Oh, okay. "... must master within the next six months." Oh, hi Angie.

Chima Mmeje:
Oh, this is a good one. I think I would say what Daniel said earlier, which is first of all, understand how these LLMs and how all these programmes work. A lot of them are open source. You can literally go online and read the patterns or the updates on how they work and maybe use... I don't even want to say use NotebookLM because you'll miss a lot of information. Read it word for word. If you have to do it slowly, one hour every day, read those patterns, and maybe then use an LLM to understand what it means for SEO. That's a good starting point.
A lot of the things they're telling us to learn, relevance engineering, a lot of other stuff. If you're learning all these things and you don't even understand how the tool works fundamentally, you already have a problem. So I think first and foremost, the first thing we need to know, understand how they work and how they impact search as we know it right now.

Jojo Furnival:
Okay. Okay. It is five o'clock, guys. So we're going to have to say goodbye. That's all we've got time for. Thank you so much for joining and for all of your awesome questions. Huge thanks to Daniel, Chima and Dixon for giving up your time and expertise today. You guys are all amazing. This was possibly the best one ever, in my humble opinion. We'll be emailing out the recording tomorrow to everyone who registered. If you missed the start, do not worry. You can catch up.
Next month's webinar is going to be an extra special workshop with Lily Grozeva on AI search auditing. So her process for how you can actually audit for all of this kind of stuff that we've been talking about today. Make sure you sign up for that one. The link is maybe going into the chat right now. Thank you again for watching and see you in the next one. Bye.