Marketers of the Universe: A digital marketing podcast

AIO: Stop chasing clicks and start building authority

• Brew Digital • Season 1 • Episode 47

Share episode

Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.

0:00 | 35:03

In this episode of Marketers of the Universe, Tom Inniss is joined by ShinRoo Chao, Senior SEO Manager, and Rich Harper, Head of Digital Marketing at Brew Digital, to dissect the rise of AI Optimisation (AIO).

As organic traffic patterns shift and AI-generated answers become the norm, we ask the burning question: Is SEO truly dead, or has it simply graduated to a more sophisticated era of brand authority?

What we cover in this episode

  • Defining AIO: We explore whether AI Optimisation is a brand-new technical discipline or an essential extension of traditional SEO that focuses on how Large Language Models (LLMs) interpret, analyse, and compose information.
  • The shift in search behaviour: With reports suggesting organic Google traffic referred to websites dropped by 33% last year in the US, we discuss if the "era of 10 blue links" is psychologically over for the consumer.
  • Optimising for LLMs: ShinRoo explains why clear structure, topical authority, and entity-based content are now more critical than simple keyword density.
  • The death of "vanity metrics": Rich argues why marketers must move away from measuring clicks in isolation and start talking the language of the board—focusing on pipeline, revenue, and brand perception.
  • Query Fan Out and content strategy: How to capture the secondary questions AI systems generate when breaking down a single user prompt into multiple sub-queries.
  • The future of content: Why "the middle is dead" and why your strategy must pivot from high-volume "churn" to high-value, authoritative contributions to the web.

Marketers of the Universe is brought to you by the clever folks at Brew Digital. We’re not your typical digital marketing agency; using an innovative approach to decision-making and collaboration, we help you create an impactful digital strategy that actually delivers results for your business.

See what we can do for you at brewdigital.com


Tom Inniss:

Hello and welcome to another episode of Marketers of the Universe. My name is Tom Innis, and I am the content marketing manager here at Brew Digital, and I'm joined today by Shinru Chao, our senior SEO manager, and Rich Harper, the head of digital marketing here at Brew Digital. Today we are looking at one of the latest hot buzzwords, AIO, AI optimization, and asking whether or not it is an evolution of SEO or if it is just something everybody's hyping up because it has AI in the title. Let's dive straight into it. So cue the music. Shinri, shall we start with what AIO actually is and whether or not you view it as a new technical discipline, or is it just a rebrand of SEO?

ShinRoo Chao:

So when you mentioned AIO, the first things I thought of is just like AIO review from Google, but there's a lot of tons just like AEO, GEO, so everything about AI search things. I think it's not simply a rebrand or just like different from ACO. It's more like an extend traditional ACO practice. So while we know the traditional ACO focuses on optimizing for like a search engine crawler that keyword ranking or rank the page, but the AI search targets more like language models that analyze, interpret, and compose information from like a lot of different uh sources. So this approach emphasizes clear structure, the right answers, and so-called topical authority. So the content uh can be easily understood and uh cited rather than just like indexed or ranked.

Tom Inniss:

So, from like an on-page perspective, do you feel that optimizing for an LLM is different from optimizing for a standard crawler?

ShinRoo Chao:

It's not like totally different. It's more like you shift the focus from one to the other. So just like we uh mentioned the content should be more structured. So structured data is actually quite important. How you use different terms as an entity and uh that those like AI search platforms understand your topic is also important. So I wouldn't say it's like totally different. You need to have a brand new uh tactic to deal with the on-page, but just kind of prioritize different things.

Tom Inniss:

Great, thank you. And Rich, I'm gonna come to you for this because this is something you've sort of been banging on about for a little while now. Um, and it's the shift away from people searching, using keywords, and moving instead to asking chatbots or even watching things on TikTok. In your mind, do you feel that that era of the 10 blue links that you see on a search engine result page is that psychologically over for the consumer?

Rich Harper:

Yeah, uh there is certainly a shift in behavior. I don't think uh we have seen the era of the 10 blue links um completely being over. Um, interestingly, um a study came out this week at um MP Digital um or Neil Patal, basically asking ChatGPT users uh whether or not people were now starting to utilize Chat GPT as as one LMM uh for search over Google. Um and what they found basically was no um still overwhelmingly um in favor. I think the numbers that we're talking like high 80%, people still use Google for search. And interestingly, from a SEO perspective, we have to consider how Google, if Google still dominates the search landscape, and although there are a growing number of let's call them searches on LMM platforms, they are still nowhere near the size of the searches that happen on a daily basis on Google. And Google trains its own AI based on its search data, um, it uses fan queries to do millions of Google searches at one point in time to then uh inform its own AI. So if if companies are not still prioritizing SEO and trying to chase a new um, you know, a new coin phrase, you know, fundamentally they're missing out big time uh in terms of of traffic.

Tom Inniss:

Well, that actually leads me nicely on to a different study that Reuters Institute uh released last week, I believe, and they found that organic Google traffic dropped by 33% last year in the US. So I was wondering whether or not they are missing out on this traffic. If Google, like Google is still the predominant search engine, sure, but I don't think we can say that it is driving traffic in that same way now. So when you see a number like 33% drop in organic traffic, do you feel that SEO is failing, or is this now the cost of doing business that companies need to accept you are going to see less traffic?

Rich Harper:

Are you saying traffic's dropping from Google to an external site?

Tom Inniss:

No. So Google is now referring 33% less traffic organically to um websites.

Rich Harper:

Yeah. So SEO is not failing in that sense. It's a change in mindset from marketers. You know, traffic is no longer the key metric, but that doesn't mean that the impact that search is having on your brand is not is not valuable. What we need to do is move away from measuring stuff in isolation, which we've always been very good at, because as marketers, we need to be able to, you know, we need to be able to have metrics to prove our value. And I think that's the the wider story is is not is SEO failing, is marketing failing at a board level, because we've always used, you know, really these vanity metrics to try and prove our worth. If we start talking the language of the board and we start talking the language of key stakeholders at that table, which are often, you know, the CEOs, the finance teams, none of these people care about site traffic. They care about the ultimate end goal, which is often pipeline revenue, that that side of the business. And we've always been, as I think, as a industry, not close enough to being being able to prove that. So the fact that traffic is down is a concern. But if we look at things a bit more holistically and a bit more from a kind of multi-channel perspective and not worry so much about the individual KPIs, search is playing its part in a in a kind of wider uh impact on a business than just they turned, they typed in this keyword, they then clicked our site, they then hit our site and got the information. Um and I think it's more of a readjustment in our minds than being too concerned about that particular drop, if that makes sense.

Tom Inniss:

Yes, absolutely. We'll come back to looking at metrics a little bit later, but I wanted to focus on the pipeline first. Um Shinru, so the data were also suggested from Reuters again that the utility content, i.e., definitions and the basic how-tos, etc., are getting hit hardest because AI can answer it instantly, basically. Um, the Google AI overviews will just answer your question and never refer you to a website. In the past, a lot of uh strategies, including our own to an extent, has revolved around um resource hub content and creating that value add, all that top-of-funnel educational content, even. If we're not getting the clicks anymore, is there any technical reason to keep producing those pages? I am I basically going to be out of a job.

ShinRoo Chao:

No, you still have your job. So uh the short answer is like we we can't stop creating top-of-funnel educational content. So it's not like you should stop it because even when user don't click, these pages still kind of trend the AI system to recognize the expertise or cite the brand and shape how answers are generated. So you still need the content to fit in the AI search platform. So technically, they are still essential for topical authority, like internal linking and the influencing higher intentionally later on, even if like click through rate is lower. So that's why they are talking about how to organize or structure your topic properly. It's kind of will be helpful for you to be incited or being uh one of the sources in the AI search platform because those type of funnel content still improve the content structure and the entities, making the topic stronger.

Tom Inniss:

So we still want to focus on that content cluster going forward, on the hope of being cited or having our content scraped by AI. Okay, that's good. Reassuring that I have a job some more for the short term at least. On the flip side, we've seen Query Fan Out uh be spoken about a lot now where AI can break one question into five sub-questions. Do you see much opportunity there for us? And how, if so, do we capture those secondary questions?

ShinRoo Chao:

Uh again, sure, yes. Uh the query fan now is kind of real opportunity, but only if the content is built and structured to capture the sub-questions, not just like like query. So for example, if a user asks, how should we structure Jira for growing thing? So the AI system will break down to like, oh, uh, the question this user asks is like, what is Jira hierarchy? How to do the epic story or task, and what problem actually uh occur at scale, and what's the backbacks practice for it, and any tool can help. So it's kind of, I think, like, oh, it's not just one question, it's like five different questions. So they will put the answer from multiple sources and to synthesize uh one response. So if we want to capture the secondary questions, we need to structure the data using like a clear question basis and the headings and answer each one directly in like a few short sentences. And uh this answer should be like easy to extract uh with supporting pages, just like internal linking is important to the pillow page and to the subtopic pages. So you can show the authority that you have on this topic. So the focus is on clarity and the structure first, uh, with the kind of a light use of structured data where it helps.

Tom Inniss:

Great. Are we gonna end up with just pages that are FAQs all the way down, which is like header answer, header answer?

ShinRoo Chao:

Uh it's more than that. Of course, FAQ is to do like uh everyone finding, not everyone, but most people find it helpful. But it's more like uh you can have like a one-pillar page to discuss the topic broadly and break down into different paragraphs, like different questions. But for those uh sub questions, you can actually expand to like a subtopic to use more supporting content. So it's kind of a triangle way. You have pillar pages, the main topic, and you have subtopic to support this pillar, and then you have subsubtopic to support the subtopic and the pillar. So it's like how you structure your content into a more organized way is very important.

Tom Inniss:

Okay, it's sounding a bit matrix webby, just like his, it's over here, it's over here, it's everywhere, and much shorter as well. Shall we talk about metrics then? I know Rich has uh a real bugbear around vanity metrics, etc. If we're seeing traffic drop, do we need to sort of forget about click-through rates now? And if AI is just scraping all of our content, amalgamating it all into um an AI overview that doesn't really send any traffic to us, are there any opportunities to actually monetize that or view that click-through rate as a win, or is it now wholly vanity and we just need to forget it?

Rich Harper:

Yeah, I guess in essence it's always been a little bit of vanity. Like we said in the in the previous section, it's about marketers proving their worth. Um search strategy, I think, is move like we said, website traffic is decreasing, it's going to continue to decrease. I think Google may say that it's not decreasing at the rate that some of those reports are suggesting. And I guess it really depends on what research you read. I I have seen Google kind of talk around about two and a half percent, or in general, search traffic is fairly stable. We look at our clients and we do see certain channels, you know, organic channels are showing a decline. Um but it's about changing that um viewpoint. So CTR is still important in a in a way, but it's about looking at search as a uh as a visibility in a brand perception tool. It's about building you know brand authority. I think you can see that behavior suggests that if a brand is mentioned in an AI overview or a snippet or an LMM, there's almost an element of trust um that forms. And then look at other related traffic, such as you know, brand searches on Google, use Google Search Console to measure if your brand is being mentioned more. Uh use Google Analytics or other analytics programs to look at whether or not direct traffic is increasing. Um if there's no click directly from the organic listing, but you're seeing increases elsewhere, it shows that you are having an impact. I think it's like we said, moving away from focused on keywords, um, start to think about more conversational search and how people are looking for you. Um do not neglect the basics. And like I said, look at it from a kind of a multi-touch point perspective. Um you know, we've always talked on this podcast about the full funnel approach, and we've always talked about ensuring that you know, when we've talked about paid media, if your brand is not strong, your paid media is going to be expensive. All of those rules apply um in this essence. Make sure your brand is strong um online uh and you will I think negate some of the declining traffic. You just need to look at it more holistically than uh than a you know a single metric.

Tom Inniss:

Yeah, no, that absolutely makes sense. Just uh quickly detouring, um when you are searching, and this is a question for both of you, when you're searching and you get an AI overview answer, how many of you are actually looking at where those where that has been sourced from and looking at the list of websites? Or do you just take it as granted that what it is says it's true now?

Rich Harper:

Um it depends on what platform I'm in. If I'm in ChatGPT, it doesn't necessarily always cite the answer, so I kind of go if the if chat GPT tells me that that's what it is, then that is what it is. Uh when I'm still a little obviously on Google I I read the AI overview or the A if I'm in AI mode, but I do often then click through uh to the source. Um because for me at the end of the day, especially some of the stuff I'm doing, if it's work-related, yeah, I I want to see the research, I want to know, and ultimately at the end of the day, like we've we've already cited in this in this podcast episode, there are a lot of different um surveys and research papers about geo, AI, SEO currently and the fact that traffic's declining, and you're gonna have to make a conscious decision of which research you follow, because there's no gospel answer to say this is exactly what's happening. And even if we go back into the world of SEO, none of the search platforms have ever told you exactly how how their ranking algorithms work because they don't want it stolen. And so that's not gonna change now. So, yeah, it depends on the platform. Going back to your question, Tom, if the source is cited, I think I would click through. If it's in an LMM, yes, I probably do take it for face value.

ShinRoo Chao:

So yeah, I'm gonna use the very typical ACO response, just like it depends. And uh like uh for example, uh, if I search in ChatGPT, I sometimes use the same prompts in Google Gemini or perplexity. I just kind of test different AI search platforms and see what answer they they give me and uh what how they structure their answer differently. So if there's of course of course, just like Rich said, sometimes ChatGPT or those AI search platforms, they not necessarily give you the source, but if they have, I will check where uh who are these sources. If I search uh AI overview, it's more like easy and clear to see where those uh information are from, and I will just I will just check all of them to see if they There's any brand or any website actually I'm familiar with. So I'll probably check those websites first. If I still can't find an answer to satisfy my curiosity, I will kind of clear out the link as well. So it's kind of makes that match depends on the situation, depends on the question. Yeah.

Tom Inniss:

Yeah. And I think that's probably the best way to take it now. Uh I I I from experience have found that taking LLMs at their word is a big risk. Um, I've used it to build my smart home and the amount of times that I've had to basically start again because it's reassured me that something works if you just type this in and then it just wipes everything, is um horrifying and perhaps indicative of why I should just learn to code rather than vibe code my way through life. But yeah, I I I feel that that is probably the way that most people are approaching it now. So it's yeah, from a from a brand and from a business perspective, I think it's it's definitely a challenge to know how effective the content you're creating is in driving traffic to your website now. But as we've said, um click-through rates perhaps don't matter as much. So, Shinroot, if Rich turned around and said, click-through rates, don't worry about them anymore, don't track them, don't worry about them, what would you focus on instead? What would be your sh like your northern star metric?

ShinRoo Chao:

Yeah, if CTR matter matters less, the dashboard probably will need to uh shift from click, which is traffic, to visibility and influence. So I think the most practical metrics are like a search visibility by topic, presence in AI overview, or like a generative answers. And of course, the brand and brand and uh entity mentioned across like a priority queries, even when no click happens. So it's more like a shift your mindset from uh only traffic clicks, there are methods, but now you need to check impression session or even the times on your websites, kind of a broad mix of uh metrics you need to look at, not just traffic. So we are actually uh measuring whether the brand is seeing, cited, or trusted in the answer set, not just whether someone wants to click through. So for example, if the old metric is click through, right, the new invisible metric could be like citation per response. Or the old metric is keyword ranking, now it will be more like an entity authority score. So just to understand how AI uh like does the AI search platform use your content, how they use it, and uh how the AI understands your brand as an expert, and how to measure the high-value traffic actually comes from those AI platforms. It's kind of need to change the metrics. Yeah.

unknown:

Yeah.

Tom Inniss:

And how easy is it to actually track things like brand authority or AI citations? Because click-through rate is pretty easy, although GA4 did their best to break it. Click-through rate is pretty straightforward. It's like somebody's hit this website, okay, cool. Whereas now you're looking almost esoterically at this is how much somebody trusts my brand. How do you actually track that?

ShinRoo Chao:

I would say it's not like still, not like the old way to track all the data, but there are more and more and more tools that you can actually track if you get cited or get as one of the source from those AI search platforms, but it just you need you need some budget to kind of uh invest in those tools. Or you can start with like a manual, manual search, use those prongs, just like try to talk with your sales department or try to think about what kind of question the potential user will ask about your product or service, what kind of question they uh uh they are looking for answers. So it's still not like 100% easy to do those tracking data, but I think that it will be uh you will get there, and it's just like you probably need to put aside a bit like a marketing budget on those tools, and uh of course you can also manually search.

Tom Inniss:

Okay. Sorry, everything is just gonna get more expensive for marketing teams. That's where we're at. The future of AI is here, everyone. Um, so Rich, industry consensus seems to be that the middle is dead, and you're now at a point where you need to either have massive authority, as you mentioned previously, or be personality driven. Does our marketing strategy, and when I say our, I mean holistically as an industry, does it need to pivot away from content volume and instead look at building famous experts?

Rich Harper:

Um, yeah, I would think aim for quality over volume, um, rather than pinpointing specifically around personalities. You know, from an up from a social perspective, yes, there's obviously a big kind of push to have personal brands within organizations and the social platforms are doing their best to push down brand pages within the algorithms in favor of personalities. But, you know, when we're talking about search, focus on high quality content, invest in that authoritative informational content. You know, do you have voices in your business, in your industry that are trustworthy? You know, are they seen as thought leaders? Try your best to make that content as unique as possible, or at least from a viewpoint, like have an opinion on stuff. You don't need to just regurgitate the same crap that everyone else is saying. You know, be if you're going to be authoritative on something, that means you have an opinion and make sure you voice that opinion, don't just, you know, uh follow the follow the crowd. Yes, from a technical perspective. You know, Shinri's already mentioned about FAQs and structured content, but let's not let's not go back 20 years and flood the internet full of crap because we think that's how the platforms are best going to retrieve our content. Um, I've already seen, I think, articles and you know, those black hat SEO techniques coming back in with people creating invisible pages just for LMM citation and stuff like that. Let's not start going back down that track. You know, the platforms will soon smart up to it if they hadn't already. They want to be driving content, you know. Google wants to be putting content in front of its users that is of high value. So it's in our best interests to be creating stuff that that is high quality. And I I think volume, volume is just spam, isn't it? We've spammed the internet for so long. Let's get out of that mindset and start thinking about what is actually driving value for our businesses and what do our customers, we've talked about this before, what do our customers need to be hearing? What do they want to be reading about and what is relevant to them? Then produce content that matches that. You don't need to write 10, 15 blogs about the same thing every month because well, believe me, one, no one cares, and two, no one's reading it because the traffic is not going to be there. So yeah, if that's one takeaway, stop producing crap, start focusing on value.

Tom Inniss:

I mean that's the title, that should be the title of the episode, but we'll we'll keep it a little bit more um podcast syndication friendly. But yeah, just thinking back on everything you've said, it's like god SEO really ruined the internet, didn't it? Sorry, Shinru, but you and your ilk, you made such a miserable experience.

Rich Harper:

Um content people ruined it, Tom.

ShinRoo Chao:

No, you should say it's not content people, it's the search engine actually ruined.

Tom Inniss:

Yeah. The monetization of the internet. We actually just got that fundamentally wrong at the start, but that is a topic for another podcast. Um, because I could talk for hours about that.

Rich Harper:

And we're probably about to make the same mistakes with AI when it comes to monetization, so we haven't learned anything.

Tom Inniss:

No, I mean we yeah, who is it? Uh ChatGPT, OpenAI, have now said we're gonna stick ads in, and that's the end of that product. Yeah.

Rich Harper:

There you go. Every social media channel. Oh no, it's free. It's every, you know, this is how we're going to communicate moving forwards, but we actually need to make money. Let's just flood it with ads and then piss everyone off.

Tom Inniss:

But it's one of the most lucrative industries in the world in terms of like margin. So you can understand why they do it, but it's yeah, it's it's it's just a bit shit, innit? Anyway, let's wrap this up before we get all existential. Um, very quick answers here. Shinri, what is one thing that SEO needs to stop doing immediately?

ShinRoo Chao:

Um, maybe just need to start optimizing purely for click or traffic. But I guess. Oh, I think most of ACO they know about this. It's more about how they tell other people or how they do the education, like uh tell other people to shift the mindset. So, yeah, that's it.

Tom Inniss:

So we've got to stop doing and then start doing as well.

ShinRoo Chao:

Yeah.

Tom Inniss:

Love it. Helpful. And then Rich, one thing that CMOs need to start paying for immediately.

Rich Harper:

Paying for?

Tom Inniss:

Yeah, they've got to spend some cash. What should they put it towards?

Rich Harper:

We drop a self-promotion or hint that they could work with Brew Digital and we will support their uh all their digital marketing needs. I'll allow it, yes. Um one thing they need to start paying for immediately. Um I think invest in proper attribution modelling. Uh if it was if it was down to me if I was a CMO and you want to know how marketing is driving influence in a business, you know, let's stop looking at last-click attribution and first-click attribution. Let's invest properly in attribution modelling that shows what is directly driving revenue from marketing activity and what is influencing revenue from marketing activity, because that will help us have more authority at board level.

Tom Inniss:

A very thoughtful answer. I like it. And finally, to come back to the top, is AIO the death of SEO or is it just its graduation? Rich, we'll start with you.

Rich Harper:

I just hate the coining of new acrins. Uh, is SEO dead? No, it's not. Um, this is the evolution of digital search behavior. Let's call it for what it is. It is still search, regardless of whether it's on an AI platform or Google or should we say Bing or any other uh available search engines? It's it's not the def, it's the evolution of search. Let's stop thing. We've done this forever, haven't we? It's like, oh, email marketing's dead, search marketing's dead, blah blah blah. If everything's dying, how are we as businesses getting in front of people? Yeah. So no, it's not the deaf. Let's stop making up new words for things that already exist. Let's just focus on what it's always been about is getting in front of the customer with the right message at the right time. Shinriu, anything to add?

ShinRoo Chao:

It's just very interesting. I have seen so many people discussing, oh, ACO is going, it's gonna be the death of ACL for the last, I don't know how many years. Everyone's talking about this. But definitely it's not the death. It's just like you said, the graduation from a transactional model. It's like I provide a link, you give me a click, to a more like authoritative authoritative model. It's like I provide the answer, you recognize my brand, something like that. So the death only applied to those low-value thing content, like the quantity things that AI can summarize easily, but for the AI search era, it's the new opportunity to own the entity search landscape, but not just the blue links.

Tom Inniss:

So, what we're hoping for is the end of endless churn and instead intelligent contributions to the web. Yeah. One can hope. One can hope. Anyway, thank you so much. Thank you so much for listening today. We love that you enjoyed our content. Uh, if you did enjoy it, please leave a like and give us a follow. And if you didn't enjoy it, send it to somebody you don't like because there's no point us both being disappointed. So um, we have been Brew Digital. Uh, I've been Tom Innis, Shinru Chow, and Rich Harper have joined me for this discussion on click-through rate. Uh, thanks very much for listening, and we will catch you in the next one. We have been Marketers of the Universe?

Rich Harper:

With Masters as a Universe coming back this year as well, can we like leverage that somehow?

Head Shot

Haydn Woods-Williams

Host