Mastering Topical Authority in a Multi-Platform Search World | Andy Chadwick
Welcome back to The Search Session podcast! I’m Gianluca Fiorelli, and I’m joined by Andy Chadwick, who describes his work as “driving brand visibility across every search platform.” Andy brings a sharp, forward-thinking perspective to how SEO is changing, so we dive into what that really means in a world of AI, fragmented search, and evolving user behavior.
In this episode, we talk about:
Ambient visibility: the idea that your brand needs to be everywhere your audience is searching—not just on Google—because AI isn’t pulling answers from just one source anymore.
The mistake of cutting top-of-funnel content: even if it gets fewer clicks, it builds topical authority and gives you something valuable to earn links from.
Topical authority and how it’s built: it isn’t a metric—it’s earned over time through content breadth, depth, strong internal linking, and recognition across the web. Andy calls this “the rising tide effect.”
Content fragmentation: A method Andy used to outrank big players by breaking down big guides into focused, cluster-driven content.
Keywords are not dead: they are still the Lego bricks of content strategy. Even if we now call them “topics,” they remain essential for understanding search demand and prioritizing content.
Andy’s “3D search intent”: it’s no longer just about whether a query is informational or transactional—it’s also about where your audience is searching.
This, and much more, in a conversation you’re warmly invited to join.
Video Chapters
Transcript
Gianluca Fiorelli: Hi, I’m Gianluca Fiorelli. Welcome back to The Search Session! Our guest today is probably someone you already know because he is the co-founder, along with another big name, Suganthan Mohanadasan. Sorry, Suga, we're friends, so forgive me if I mispronounce your name. So, my guest is also the co-founder of a tool that I hope many of you are using already, which is Keyword Insights.
Also, he’s the co-founder of an agency called Snippet Digital—an SEO agency. And remember, not a GEO agency or something else.
And something a bit more personal: he’s Scottish, even though he’s living in England now. I’ve read quite a few things online, and he’s one of those classic SEOs who learned SEO by doing it. In fact, if I’m not wrong, he taught himself how to do SEO in his early twenties, when he was basically running a roller shutter business, if I’m not mistaken.
He considers himself a perfectionist—but he’s been able to do something I haven’t quite managed myself: delegate. He’s a remote work advocate, and that’s actually one of the reasons he co-created Keyword Insights. He’s also really passionate about using creativity to achieve goals—especially and not only in the SEO world.
Our guest today is Andy Chadwick. How are you doing?
Andy Chadwick: Hello, mate. Good, good—thank you. Thanks for having me on. By the way, I struggle to pronounce Suganthan’s surname as well, so don’t worry about it. And I’ve been working with him for five years!
Gianluca Fiorelli: That’s why we all call him Suga.
Andy Chadwick: Easier.
Gianluca Fiorelli: It’s like you're in Spain with my name, which is weird—because Gianluca—you might think that Spanish and Italian have somehow the same name roots, due to the Latin and things like that.
In Spanish, there are a lot of names that start with Juan and then another name after it—like it is Juanluca but no Gianluca. So I end up being called Jean-Luc, like Jean-Luc Picard or Juan Luca, which is a composite. I’ve even been called San Luca, which is actually the name of a little town in Andalucia, Sanlúcar de Barrameda.
And when I go someplace and I know nobody’s going to be able to pronounce my name, I just call myself Luca, especially if I’m going to somewhere like a Starbucks or any kind of coffee shop.
Andy Chadwick: Makes sense.
Gianluca Fiorelli: Andy, your name is definitely simpler—much easier to pronounce. I hope Chadwick is the correct pronunciation.
Andy Chadwick: Yes, yes, it is.
How SEO Is Treating Us in 2025
Gianluca Fiorelli: Great—I don’t want to spend an hour mispronouncing your name! So, let’s start with a classic question I ask all my guests: “How is SEO treating you lately, with all these confusing times we’re living in?”
Andy Chadwick: Oh, it’s still great for me, to be honest. I’ve always thought—and I keep telling people—I feel like SEO is now what it should have been 10 years ago.
For far too long... I mean, don’t get me wrong, I loved it when it was easy—but it was too easy. And that’s what led to just a load of crap being put out on the internet. Now it’s not so easy, and now it’s a lot more varied. I actually find it more interesting. It’s not just about optimizing for search engines in the traditional sense anymore—it’s about optimizing across all these different platforms.
And to be honest, we started doing that before AI became a thing. We were already moving in that direction when TikTok started taking off. So we've always had this mantra: “Be where your user is searching.” That didn’t make us a traditional SEO agency—we were doing search wherever.
And now that everyone else is leaning into that, we’re already ahead of the game. So yeah, SEO is treating us very well. In fact, we’re probably doing better now as an agency than we were two years ago—in terms of the offerings we have and, I guess, the angles we can bring to clients.
The hard part actually isn’t doing the work—we were always doing the work. The hard part is reframing it to key stakeholders. Two or three years ago, SEO used to be: you’ve got paid PPC advertising, which was a performance channel. You give me five quid, I’ll turn it into twenty, and I give you a return on ad spend figure. I could give you a metric that tells you: “You put money in, here’s how much money comes out.” SEO was kind of like that—it was part performance, part brand building. It sort of sat between the two.
Now, I’d say it’s leaning more and more toward the brand-building side of things, which is a lot harder to justify in terms of cost—because you can’t give them a clear money-in, money-out figure as easily.
But it becomes a case of reframing it, like: “Well, hey—Mr. Client, or Mrs. Client—you’re very happy to pay £5,000 to put up a billboard or advertise on TV. That’s brand building. So you should be just as happy to invest in SEO, if you reframe it that way as well.”
So I think that’s the hardest bit: reframing and managing expectations. But in terms of the actual work—yeah, I’m quite enjoying it. And we’re doing well off the back of everything that’s happening.
Gianluca Fiorelli: Yes, I completely agree with you. In fact, I think—I mean, it was easier until a few years ago. Let’s say, maybe 18 months ago, when all this stuff started to burst. More than a year ago, actually.
Because if we think about it, the AI Overviews, for instance—which is something that really impacted SEO—that’s only about a year old. But it already feels like it’s been with us for many, many years.
And yes, maybe things were easier back then. So we were also trying to find a personal challenge. Probably you, as an agency, and I, as a consultant, were looking for projects that were not only interesting, but also fulfilling—for the kind of challenges they were giving us.
Because otherwise—sure, every business, every website has its own story, and the solution is never the same for all the websites—but somehow, it was starting to feel the same. You know, a bit like Charlie Chaplin in Modern Times: “Okay, I start by doing this, then I do that, and then this...”
I’ve been lucky enough—like you said—that many of the things which are considered super important now, because of LLMs, I was already doing. And I’m not talking about all the classic, super-silly SEO tips that some AI bros are now sharing with everyone, asking people to follow them to get a checklist.
I mean things like topical authority, clustering, all that kind of stuff. So in that sense, I think we share a common ground—a common journey through these types of solutions.
“Ambient Visibility” in SEO
Gianluca Fiorelli: You were talking about how to present this change in terms of how to measure things—and I think that’s probably the biggest challenge. And actually, on this topic, I want to bring up something you shared a couple of months ago on LinkedIn. I have it in front of me: “Organic clicks are fading. As they vanish, is “Ambient Visibility” the key to your brand's AI dominance?” Can you explain to me—and to everyone watching—what you mean by ambient visibility?
Andy Chadwick: It’s just coming back to this idea of being everywhere. Where traditionally, you might’ve just put a load of money into being on a search engine—you’d write a blog, maybe 5,000 words, covering lots of different subheadings and subtopics. I'm now talking about this idea of ambient visibility: breaking that up, splitting it out, creating loads of guides targeting overlapping—or somewhat overlapping—terms.
Then, pushing that content on YouTube, pushing it on social, spreading it across all these different platforms. Making sure that, if you’ve got a point of view, you're also pushing it somewhere like Reddit. Because what AI models are doing now is not just looking at one thing. They’re pulling from multiple sources—synthesizing all of that. They’re pulling from YouTube, Reddit, ten different websites, and then combining that information before spitting out their answers. And so, you want to be ambiently visible across all of it.
And it’s hard—especially for a small business. Because traditionally, they could’ve just stuck out a blog 500 quid, job done. Now we’re telling them: “You need to be on Reddit, you need to be on TikTok, you need to be on YouTube. You still need to write those blogs, those top-of-the-funnel and middle-of-the-funnel guides. You still need to do all of that. But you’re competing against the big players.”
What I’d say to that is that although AI has made it so you now have to be everywhere, it’s also been the great leveler—because it should now be a lot quicker to produce that content.
I know quite a few companies that have simply stopped writing top-of-the-funnel, blog-style guides. Their thinking is: “Why bother? AI will just steal that click anyway.” I can tell you—if you’re not doing it, your competitors are. And it still gets cited, it still brings some referral traffic. Not a huge amount, but still some.
And it also gives a better context of who you are, which is really important when an AI, an LLM, or even a search engine is trying to understand your brand. The more surrounding content you’ve got, the clearer that picture becomes.
So yes, you still need to produce that content. But the ROI of producing it has gone down—and at the same time, the cost of producing it should have gone down too. You should be using AI—not to write the whole thing—because just churning out guides with AI isn’t the way to do it—but to structure it.
I recently did a Moz Whiteboard Friday where I spoke about how to use AI to write a blog the right way. And then you can use that same AI output to push content across all the other channels. You can use it to draft your TikTok script, your YouTube post, and so on. So yes, you need to be in more places, and you might think it’s going to cost a lot more—but AI has leveled the playing field. It’s now much cheaper to distribute content across all those places.
Gianluca Fiorelli: Yes, you are substantially talking about embracing omnichannel as a content strategy. And I think you are totally right. It was, somehow, a strategy that was needed even before the impact of ChatGPT, Perplexity, Claude, and so on. We were already seeing how Google was filling the SERPs with features of every type. So if you wanted to gain visibility in Google, you needed to be present in all of these features.
So, if a feature is an image box and you’re a website that is strongly visual—in the sense that people look more at your product images than the descriptions—then you need to be really strong in visual search. That means optimizing for Google Lens, for image search itself.
And if there’s video, then you have to produce video—or, eventually, work with creators in order to have your brand present in their videos.
Gianluca Fiorelli: So, I totally agree with this. And I also completely agree on how AI, as a tool, can leverage ability. Because now everybody has access to the same tool. It’s not about the cost or price of the tool—it’s about the ability to use it, and the creativity in using it, that makes a brand stand out.
We’ve already seen examples of very big brands using AI and doing what, in Spanish, we call a chapuza—basically, a really bad job with it.
I think you’re right: the real problem is how to explain this to clients, and make them embrace the needed change, instead of just giving five minutes of uncertainty and then saying, “But we can do these things.”
I also agree about the fact that some websites, some companies, are completely cutting the top of the funnel. I also think that’s a mistake. Not only because, without this kind of content, you’re far less visible—since 80% of searches, across every platform, not just Google, are informational. If you’re not creating informational content, then you’re not visible for 80% of the opportunities.
But also because—and here I want to introduce another topic you know very well—informational content is essential for creating topical authority for a brand.
Topical Authority: What It Is (and Isn’t)
Gianluca Fiorelli: So, talking about topical authority: what do you think people still don’t get, or don’t completely understand, about this concept?
Andy Chadwick: I don’t actually know. That would be a really good question for us to ask our own customers because Keyword Insights is obviously framed as a platform to help people become a topical authority. And I’ve never stopped to ask the customers what part of it they don’t understand—which might actually make the tool a lot better.
I think it’s a hard thing to conceptualize because it’s quite a nebulous term. Like, what is it exactly? Some people have tried to create a metric for it, but there isn’t really a clear metric. You don’t just have one unit of topical authority or 100 units of topical authority. It’s a bit vague, and it’s definitely hard to quantify. So I guess the real question is: what does it actually mean to have it? And that’s where the struggle lies. That’s a good question—and one we should probably be asking our own customers. What do you think?
Gianluca Fiorelli: Well, I think that when I’m talking to a client, I usually use an allegory to explain it. I say: “If you’re feeling some pain and you go to a doctor—let’s say it’s pain near your heart, radiating through the shoulder—who are you going to consult? A cardiologist or a physiotherapist?”
The person with the expertise and experience—because they’ve spent years solving problems like this for other patients—is the one who has topical authority on how to resolve it. In this case, the cardiologist.
In terms of websites, when I explain it to clients, I say: “If you’re a site offering tours, vacations, and hotel rooms in Tuscany, then you should demonstrate that you cover all the things surrounding the topic of Tuscany.”
In the niche of tourism and travel, you show—through your content, your shorts, your images—that you have expertise. So, the classic concept of E-E-A-T is essentially what builds up topical authority for a brand.
When I think about the machine side of things, topical authority for me is really more about semantics. If a machine like Google—or even a machine like ChatGPT, though I’m not so sure that ChatGPT or LLMs have any real consideration of topical authority the way Google does—well, in Google’s terms it means this: if you’re a website selling travel to Tuscany and you’re creating content about Tuscany, it’s not just about producing that content. It’s also about whether that content is recognized through mentions, citations, and other signals that can be found outside your own website.
For instance, maybe on Reddit, where people cite a video you’ve made, or on Instagram. Now that Google is also showing professional Instagram content, that’s important news for me. Maybe they’re not counting the likes directly, but they are seeing the engagement on those videos—and that matters.
Say you’ve done a short on YouTube, it’s performing well, and so on. If all these things—external signals, internal linking, and semantic consistency inside your website—are present, then you can eventually consider yourself to have topical authority on that subject.
I think there’s another factor that’s important, also for LLMs: consistency and recency. If you have a consistent history of creating content about a topic, and it aligns with all the other potential factors, then that adds another layer. I think this is what can truly constitute topical authority.
Andy Chadwick: Yes. For me, if I had to define it, I would call it the rising tide effect. If you have two websites—and I’ve seen this anecdotally—let’s say they’re exactly the same: they sell the same product, they look the same, everything’s the same. Both sell suitcases.
But one of the websites also talks about wider travel topics—travel tips, guides, that kind of thing—and has all the internal linking that goes with it. What we tend to see is that the site with the wider net of content tends to perform better.
I think this ties back to that nebulous idea of topical authority. The internal links, the way Google crawls and understands the site—it grows more and more confident that this is a travel website.
And secondly, links still matter. By building this top-of-funnel content, you can more safely build links to those informational pages—and then internally link from those to your product pages, which then rank your products. Because obviously, if you’re a suitcase-selling website, you want to sell suitcases. You don’t just want people landing on your top-of-funnel content.
So even if that top-of-funnel content doesn’t seem to bring an immediate benefit—even if you’re saying, “Well, it’s not driving any traffic”—I’m pretty sure, and I can almost guarantee, that it is helping indirectly through this idea of topical authority.
But even if you don’t believe that, you can still use it to build links—which is your first line of defense. You can’t go out and build a load of links directly to a red suitcase page, because that looks really spammy. But you can build a load of links to an article like “How to Choose the Perfect Colored Suitcase,” and then internally link that to your red suitcase page. That doesn’t look spammy. And that’s another big reason for making top-of-funnel content.
Gianluca Fiorelli: Yes, yes, totally. And now with the rise of LLMs—not just GPT—I’m thinking about how, in Southern Europe, in Spain, Italy, and even France, we still don’t have AI Mode. But in the UK, you already have them, and our friends in the US started to see them even earlier.
So with AI Mode and with AI Overviews, this concept of creating or formatting content in chunks is also rising. In a recent conversation I recorded with Ian Lurie, he was saying that the concept of chunks may feel new, but it really isn’t, because the way people remember information is in chunks, not as entire documents, which is true.
Gianluca Fiorelli: And I think that informational bits can also be placed inside other types of content—pages or documents that aren’t 100% informational.
You were talking about, for instance, the “how to choose the perfect red dress” guide. But on the product page for a specific red dress, you can also create FAQs specific to the product. So you can go even further—down to a very long, conversational, query-type situation.
And now, from a user’s perspective, Keyword Insights has recently introduced several new features. One of them is the ability to create a personalized context around the projects a client is building with your tool.
Keyword Insights: What’s New and What’s Next
Gianluca Fiorelli: So here’s a bit of a backstage question: how are you rethinking Keyword Insights for more specific uses in LLM content creation?
I know Keyword Insights already uses AI within the tool. But I’m thinking more broadly. Clustering, for example, has a lot of AI behind it, and while it’s usually framed as a “classic SEO” feature, I think it’s also really important for LLMs.
And for all these cases—where maybe a query-fan-out doesn’t deserve a standalone piece of content but could be a chunk inside another piece—can Keyword Insights be used to take advantage of that?
Andy Chadwick: Yes, I think you hit the nail on the head. The traditional way Keyword Insights is used for optimizing for AI now is that it helps you become a topic authority—and that’s the big message right now.
You can pull all the keywords from your competitors, group them in Keyword Insights, and it will tell you every single page you need to create. Then, with one click, you can generate each piece of content you need. That process helps you build topical authority, which in turn helps you get visibility.
Now, in answer to your question, I’ve got to be careful what I say. I spoke to Sugan just before this call, and he said, “Don’t give too much away.” But we do have new features coming very soon. As you pointed out, they’re almost the reverse of the query fan-out. It pulls all those queries in, and it will help you write the articles in a way that ensures you’re ticking all the boxes. And ultimately, it will track to make sure that you’ve created this piece of content using Keyword Insights. Then, we suggest you answer these query fan-out questions.Click this button, we’ll make it, and then you can track it.
I think I haven’t given too much away—but that’s all coming fairly soon, and it’s really exciting. That’s the direction we’re going with it.
Gianluca Fiorelli: Cool. Yes, it looks exciting, it looks promising.
Content Fragmentation Strategy
Gianluca Fiorelli: We’re talking about content clustering, topical apps, and content apps—like chunking. Two years ago, you were already talking about something you called “content fragmentation.” What did you mean by content fragmentation? Although it’s two years old, I still feel it’s very relevant.
Andy Chadwick: Yes, it’s a technique I used to use to outrank big guides like Forbes. For example, I had a client in the crypto space, and Forbes dominates cryptocurrency—it ranks for nearly everything. So we were like, “Okay, how can we beat this massive company, with all its authority, for all these terms?”
What you do is this: let’s say you type in “what is Bitcoin.” I don’t know if Forbes still ranks for it, but let’s say they do, and their article comes up first. If you take that URL and drop it into Ahrefs, Ahrefs will tell you, let’s say, that page gets 10,000 visitors a month—though it’s probably more like 100,000.
And it’s made up of all these different keywords that it ranks for. “What is Bitcoin” is one of them, but so is “how much is Bitcoin”, and “where do I buy Bitcoin”, and all those different queries.
What’s interesting is, it doesn’t rank number one for all those queries. It might rank number one for “what is Bitcoin,” because that’s the main query, but only number nine for “how much is Bitcoin.” It ranks number seven for “where can I buy Bitcoin.” It ranks number eleven for “how do you mine Bitcoin.”
It ranks across 200 or 300 different keywords, and where Ahrefs gets that number for organic traffic from is all of them. But it only ranks number one for a small group. Then it ranks number seven for another group, number ten for another group, and number three for another group. So what you’re really seeing is, okay, Forbes is a big player, but it’s ranking for so many queries across so many results. It’s only holding those positions because of its authority.
So what you can do is grab all those keywords, drop them into something like Keyword Insights, and it will analyze the search results and group the keywords based on what’s ranking. And so, okay, maybe Forbes ranks on the first page for all those keywords. But actually, for the other 299 keywords, there are different guides bouncing in and out—super specialized guides on whatever the exact question is.
So maybe Forbes’s ultimate 5,000-word guide ranks for all these keywords. But at the same time, there’s a specialized guide on “where to buy Bitcoin” that ranks, and another specialized guide on “what is Bitcoin” that ranks.
This idea of fragmentation is pulling all the keywords that Forbes ranks for, dumping them into a clustering tool, which then analyzes the search results and groups all the clusters together—basically telling you how many separate pages you can split that one big guide into.
Then you go and create all that content. You’re not going to outrank Forbes with just one piece of content, but you might outrank them with ten—each targeting a different query. And that’s what we use to compete against companies like that.
So, the idea of content fragmentation is splitting these big authority pieces into ten, twenty, or however many pieces they may be.
Gianluca Fiorelli: Cool. And essentially, you're using content fragmentation as a complementary way. Because with all this fragmented content, you're essentially creating a cluster. Again, we're coming back to clustering content around a topical theme.
Andy Chadwick: Yes, exactly.
Gianluca Fiorelli: That’s really interesting.
Are Keywords Dead?
Gianluca Fiorelli: And now, another question—it may sound a little weird, but in the last month, there’s been this constant tantrum—let’s call it a tantrum—around the idea that "keywords are dead." How does that feel as a recurring trope to someone who co-founded a tool called Keyword Insights? What do you think is the role of keywords now, especially since we so often talk about entities and topics? Where can we refrain keywords?
Andy Chadwick: Anyone who says that, I’d challenge them to not use a keyword research tool ever again and see how their SEO or their organic traffic holds up. They won’t do it.
Ultimately, people are searching in long phrases now. We're calling it conversational search, or there's key phrase search, or people aren’t searching in entities. But we're talking about conversational search. “Keywords are dead, conversational search is the way forward” is what I'm hearing a lot.
And when you look at these conversational searches—I’ll just call them longer queries, because that’s what it is. The query going into the search engine—well, the LLM—or even the search engine now, if you're using AI Mode or whatever: 'What's the difference between Nike shoes and Reebok shoes, and which one would be better for a 5K run?' Okay, that's a really long query. But the important thing is to see it for what it is: it's like a Lego house. But really, each one of those long queries is built from Lego bricks. It’s made up of a number of keywords, each of which independently has its own search volume. And if you understand what that fan-out looks like—so one is “Nike shoes” and you want to write what a Nike shoe is, one is “Reebok shoes”, one is “5K”, “how to win a 5K”—this is what we're saying about becoming a topic authority.
If you cover all of that content—what is a Reebok shoe, what is a Nike shoe, what is a 5K, what's a good time to do a 5K, which shoes would you recommend doing a 5K in—you write all of that and you interlink them together, it's gonna synthesize from you anyway.
And the thing that’s guiding you, writing those topics, is keywords. So it’s just—you still need that data. You still need the keywords. Because we can't possibly start to get search volume on key phrases, because you’ll never get any. So you still need to break it down. Okay, let’s call “keywords topics”.
You still need the search volume at the top—and search volume’s never been accurate anyway, right? I just use it as a marker. Like, if the word “Nike” has a thousand and “Reebok” has 500, I don’t literally take it as, 'A thousand people a month search for “Nike” and only 500 search for “Reebok”. I take it as: “Nike” gets twice as many searches as “Reebok”.
So, you still need the keywords. We’ll call them topics now if you want, but they’re still keywords—to understand the search volumes, to understand the content you essentially need to be writing, or at least prioritizing. I think you should write all of it, but prioritizing, certainly.
So keywords very much have a role in it. And if anyone says they don’t, I would dare them to get rid of all their keyword research tools and just start producing content—or prioritizing content—without them. I think I’ll get my results a lot quicker by still using keyword research data.
Gianluca Fiorelli: Yes. I think that sometimes I’ve been personally misunderstood, because I always say I don’t like keyword research. Personally, I find it very boring. But really, I don’t like certain types of keyword research—like putting in a keyword or clicking on 'Keyword Magic' in SEMrush, for instance, and then retrieving a million keywords spit out by the tool. That, to me, is very silly. And then trying to create content targeting just one specific keyword? I think that’s really 2005, maybe 2010.
Andy Chadwick: Sure.
Gianluca Fiorelli: I’ve always considered that keywords are substantially—an now I’m using another metaphor—the entity is a concept. You cannot even explain it very well to clients. Like, what is an entity? 'Well, it’s a thing that… blah blah blah.' But keywords are, essentially, the words that represent parts of that entity. They’re like the atoms of a bigger thing—I don’t even know what to call it. They’re essential in order to understand the whole. And your example was really good, because that conversational query—you can split it up.
And I was using this kind of split up. When someone was asking me, “Okay, now what keywords do we need to target?” instead of explaining, “We have to target this, and this, and this, and this,” I would say, “Look at the title tag I’m proposing to you. Let’s see all the meaningful combinations of the words that we have in that title tag. These are our keywords.”
Because obviously, the classic title tag—“hotel in Ibiza”—is easy. But something like “exclusive hotel, cheap hotel in Ibiza offers 2025”—it contains “hotel in Ibiza,” “Ibiza offers 2025,” “cheap hotel Ibiza,” “exclusive hotel Ibiza.” So it’s a big composition.
Another thing about the way I was always proposing it—and the way I always hoped a tool would finally be able to fulfill—in order to help me avoid manual work, was that Google is already suggesting to us the queries and the keywords we have to target. Because all the query refinements that it does, all the related searches, contain keywords. The “People Also Ask”—you can synthesize keywords from that, too, and so on.
And now, also the query fan-out, you can synthesize the query fan-out into a few keyword sets. So you have your keywords. Using Google, you can do your keyword search, and then you can go to a tool like Keyword Insights and start doing the more interesting part—which is not really keyword research, but how to turn this mass of keywords into something logical, something useful.
Mining Reddit for Real Questions
Gianluca Fiorelli: But talking about that, there’s a little tool—which is free—that you created a few years ago, which is Reddit Insights. It’s kind of the little brother of Keyword Insights. And I think it’s even more important as a tool because of how much LLMs are substantially relying on Reddit as a source.
So, do you have any interesting case studies—maybe from your work in an agency—of using Keyword Insights that you can share with our viewers and listeners?
Andy Chadwick: Yes, we’ve actually had to take Reddit Insights down for some legal reasons, which I won’t get into at the moment. But essentially, the concept is this—and it still works really well—you’ve got to put yourself in the mind of someone who’s going to ask a question on Reddit.
They go on Reddit mainly because they want firsthand experience—but also because they haven’t found a decent enough guide elsewhere online that answers their question. And that’s probably why they’ve turned to Reddit.
And then it becomes a self-fulfilling cycle, right? We as SEOs and content marketers—we all have access to the same keyword research tools. So we’re all basically going after the same keywords. Sure, one person might be slightly better than another at adding filters or cutting through the data, but we’re still all targeting the same stuff.
Meanwhile, we’re not writing any of the content people on Reddit are actually asking for—because the questions they’re asking are so long and conversational, they’re not in the keyword research tools. So no one’s writing them. And it creates this little chain that goes on.
If you go on a Reddit thread and scroll long enough, you start to see the same question come up again and again. Like, if you’re on one about skincare or beauty, you’ll see the same skincare question asked maybe 20 times. If you go on one about plumbing, you’ll see the same question—plumbers asking about a certain tool—a thousand times.
We had a skincare client who came to us, a brand-new venture-capital-backed company. So, loads of money, but no authority—brand new website. And they were like, “How do we insert ourselves into this conversation? It’s incredibly competitive—it’s moisturizers, it’s skincare. How do we even begin to compete on that?”
So what we did—and this is where we used the Reddit API—we pulled all the questions from the relevant threads: skincare beauty, there were loads of threads related to skincare. And then we pulled all of those questions, and when you cluster them together, you actually start to see that it’s the same question being asked. I mean, you can spot it just by scrolling anyway, but this was a very quick way of grabbing all the questions relevant to your niche, clustering them, and seeing that these aren’t zero-volume keywords, it’s just the same question being asked a hundred different times—all with zero volume. Well, it’s not really zero volume; it’s just people wording it differently, and the keyword research tools aren’t picking them up.
We made a whole content strategy around answering those questions. We wrote it from a first-person point of view so it would resonate more. It wasn’t written in second person, like “you need to do this”. It was written more like, “I got out of bed and this is what I found helped.” Because that’s what people trust a lot more now.
We wrote all that content, and I think we went from nothing to about 20–30,000 visitors a month in around three months—which, in the moisturizing industry, is really good going. So there’s a really solid way to use Reddit plus keyword clustering to get some great insights.
Gianluca Fiorelli: Yes, and if I have to give a suggestion too, it’s that the harder part is understanding what are the good subreddits to use as a source. In this case, I think another tool—Rand Fishkin’s SparkToro — is a very good one for understanding what the potential subreddits are. Not just the classic big ones, like the SEO subreddit in our niche, but also what SparkToro calls the “gems”—those smaller subreddits that might not be huge but have a lot of engagement. That’s where you can probably find fresh information.
And another thing that I think is really interesting—and I think you’ll agree with me—is analyzing trends very well. Reddit trends, classic Google Trends… and now, finally, we can retrieve that data more easily, thanks to Google. Not that there weren’t already unofficial ways to do it, of course. And Pinterest is a great source—it has, substantially, all the same trend signals as platforms like TikTok trends, and so on.
And I think this is really important because, considering the level of personalization we’re going to experience even more in search results—and it’s already clear in LLMs—the earlier we are in targeting trends, the better it is for our clients. It helps them pop up and enter into that loop of memory for people searching for those things.
So, have you had any experiences working with trends beyond the classic newsjacking tactics? How do you use trends, for instance, with your clients?
Andy Chadwick: Well, we haven’t had many clients who’ve really needed to jump onto it. I mean, the keyword discovery part of our tool—which, to be honest, still leaves a lot to be desired; it’s not the best keyword research tool out there, to be blunt—but it does something a bit different compared to other keyword research tools and that is the fact that it pulls from Quora, People Also Ask, and… oh, it’s slipped my mind now—the other Google thing that shows up at the bottom. I’ve had a total brain blank.
Gianluca Fiorelli: People Also Search For.
Andy Chadwick: Yes, People Also Search For. So, it can actually pick up trends before the traditional keyword research tools do. Because keyword research tools inherently rely on historical data, right? Whereas People Also Ask and People Also Search For change depending on what’s happening that week.
We had one example where there was a cucumber shortage in Iceland a few months ago. They completely ran out of cucumbers in the whole country. And if you used our keyword research tool, you could see it coming to life. When you typed in the keyword “cucumber,” most of the keywords we pulled back were things like “cucumber recipe TikTok.”
And the reason for the shortage? A cucumber trend had gone viral in Iceland. So we picked up on all those keywords—well, I say “before”… no other keyword research tool ever picked them up, because the trend came and went too quickly. But we caught it.
To be honest, the main way we really use trends—as an agency—is in reactive PR-type stuff. So, if we see something getting big in the news or somewhere, we’ll jump on it and spin up some kind of funny campaign about it.
Oh—we’ve seen it work with certain clients. We had a dining table client, for example, and we noticed a specific Scandinavian trend—the “Scandi” trend—starting to emerge. We picked that up through 'People Also Ask', and we suggested they start offering that as a product. Just put a landing page out there, see what happens, see if there’s interest. And if it made sense, then build it. And actually, that ended up being one of their biggest sellers for about two years. So yeah, we’ve definitely seen results from it. But most of the time, we’re using it in more of a reactive PR way.
Gianluca Fiorelli: Yeah, I think trends can be really interesting—not just for those sudden “out of nowhere” trends, which sometimes are really curious.
For instance, I remember a case with a client in the travel industry. All of a sudden, I started to see search trends for a small, relatively unknown town in Sardinia, popping up in France. It went from zero to 100 in just a few days. Maybe you even know about it, because it made it into the British newspapers. It was this town that, in order to avoid disappearing due to having no more inhabitants, was selling houses for one euro. And suddenly it went super viral—everyone was searching for it.
But apart from those kinds of cases, I think trends are possibly one of the best ways to understand not just seasonality in general, but the peaks of certain types of seasonality. It’s a great way to see when the perceived search intent, from Google’s point of view, may shift—from purely informational to commercial, or even transactional, and so on. I think that’s quite an interesting way to use trends.
The Future of Search: Fragmentation & 3D Intent
Gianluca Fiorelli: And you’re definitely one of the people who has a strong perspective—because of your experience with Keyword Insights and all the research that you and Sugan are doing to improve the tool—a good idea, a good vision, maybe not perfectly clear, nobody has it, about how search is now. But what does your gut tell you about what’s coming to be in the future?
Everyone’s saying that AI Mode is going to become the new search. I even wrote a long post about it. But now I’m seeing Web Guide, and I find Web Guide great. So what surface do you think we’ll need to work on for our clients?
Andy Chadwick: I think it’s going to be incredibly fragmented. I just can’t see commercial queries—by that I mean transactional queries—being fully replaced by chatbots.
The research stage, yes. I think that will shift more toward LLMs, so that will change, like if I’m researching, for example, “Which are the best trainers to buy, this or that?” But I think that ultimately, the final search will still happen in traditional search. It’ll still be things like “Nike trainers” or “blue dress” or something like that. I still see those kinds of queries—at least for the next five to ten years—staying on traditional search.
That’s because there are a lot of people who are older. I speak to a lot of people and have gathered a lot of interview data, and many of them haven’t even downloaded an LLM app to their phone. We live in a bit of a microcosm—because we’re at the forefront of this, we use it, we’re early adopters—so we naturally assume everyone else will be too. But I don’t think that’s the case.
So I do think SEO will still be important. Ranking on search engines will still matter. Google will probably start losing some market share, but it won’t be overnight. I don’t even think it’ll be within five years—it’ll be very slow. And by market share, I mean the traditional ten blue links with the ads and all of that. For commercial and transactional queries, I think change will be gradual.
But overall, search will just become more fragmented. That’s what that post was talking about—we need to show up on TikTok for some queries, on Pinterest for others. Like, for “bathroom ideas,” I wouldn’t waste time writing a blog post. I’d make a Pinterest board about bathroom ideas. Or something like “cool bathroom gadgets,” or “where to go in London” are searches that are actually going more towards TikTok.
People want to search TikTok for things like “bars in London.” If they’re buying a blue dress, red dress, or wedding dress—that’s still more likely on Google. If they’re comparing one product against another, or asking, “I’m going for a 5K, what should I run in?”—that leans more toward an LLM. So it’s really about understanding what I call “3D intent.”
So we have this 2D intent: “Is this query transactional, commercial, informational, or navigational?” Those are the stereotypical categories of intent. But this idea of 3D intent adds another layer. It’s asking: “Okay, yes, the query is commercial—but where is our audience?”
For example, if it’s about pension schemes or insurance for older people, I think those commercial terms will stay on Google—and even comparison searches will still happen there. But if we’re comparing the next iPhone, that might move to a different platform entirely, because of demographics and the age gap.
So it’s this idea of adding a 3D level of intent on top of the classic categories, which I find really interesting. And that’s the direction I think search is going.
It also aligns with our branding—we’ve changed it to: “We’ll get you found, wherever that search may be.” Because it’s not just Google anymore—it’s everywhere. But I do think Google will still matter, at least for the next five to ten years.
Gianluca Fiorelli: Yes, I agree. And talking about brand—but also about this sort of confusion between “brand” and “branding.” Branding is the dress: the colors, the logo, the font. But the brand is what really matters. The brand is owning your unique voice and being able to communicate, with that voice, the right message to your audience.
In Italy, I was talking with my friend Giorgio Taverniti about this concept. What you call “3D,” we call “liquid search.” In the sense that, like water, it flows from one place to another. The searcher—the search journey—flows from one surface to another surface. Before, everything was substantially owned by Google. But now it’s more like 25 or 30 years ago, when we had more search engines and Google was just at the beginning.. In Italy, we had Virgilio, Libero, and other search engines.
The Fireside Questionnaire
Gianluca Fiorelli: Let’s talk about you now. You’re Scottish, and I really like your personal website, especially your About page. I like the photo you shared there—you said you published it in order to help us feel relaxed, while also evoking nostalgia for you. What do you miss about Scotland? Even though you don’t live so far away, what are your memories of being Scottish?
Andy Chadwick: My whole family is pretty much from Scotland. And my website really needs updating—but I was born there. My parents moved around a lot, so funnily enough, I actually lived in Italy for four years, in Naples. I also lived in California and in Australia. I moved around quite a bit. Then we came back to Scotland for a few years.
For me, it just feels… do you know when something just feels like home? I live in London now, but I’m from a little town called Lossiemouth, which is really far north. You go to Edinburgh, then keep driving another four hours up the coast. And I had the perfect balance growing up there: just a five-minute walk to the beach, and Lossiemouth beaches there are stunning—they’re not like the ones on the east coast of England, which is all gross. These were blue waters, golden sandy beaches. A little bit cold—but lovely beaches. So I had a five-minute walk to the beach and a 35-minute drive to the mountains. So I had the best of both worlds.
You could surf in the summer and then ski or snowboard in the winter. It was fantastic. And sadly, the last of my family who lives up Scotland has either moved away or passed away. And so, there's not much reason for me to go back there anymore, which is sad. We used to go there every Christmas. Even when we lived in Italy, we'd go out for Christmas. So yes, it feels like home.
I had a stag do or a bachelor party on the weekend just gone, which is why I'm a little bit tired, still recovering from that. And my friends took me to Loch Lomond, which is the big lake near Glasgow, a famous song about Loch Lomond, and it was really nice to be back in Scotland there. But yes, it feels like home. I like whiskey, I like shortbread, it just feels very comforting.
Gianluca Fiorelli: I have very fond memories of Loch Lomond because when I was a kid, we traveled to Scotland by car—a really old Fiat—with my parents and my brother. We stayed in a bed and breakfast, and I remember it so well. And you were saying you lived in Italy—that’s a personal curiosity I have.
Andy Chadwick: Yes, in Naples. I lived in Naples for four years.
Gianluca Fiorelli: Cool.
Andy Chadwick: Yes, I used to speak fluent Italian, but I can’t remember it anymore.
Gianluca Fiorelli: Maybe we can organize a private lesson.
Andy Chadwick: Yes, I went to school in Italy. I moved there when I was three, and I left when I was seven or eight. Back then, I could speak fluent Italian. I even used to translate for my parents.
Gianluca Fiorelli: Cool, cool. And when you were a kid—or even a teenager—I don’t think you were thinking, “I want to be a marketer.” Obviously not an SEO, because SEO wasn’t really a job at the time. So what did the younger Andy Chadwick want to be?
Andy Chadwick: I always wanted to be a pilot. But I didn’t have the eyesight for it. I mean, I’ve had laser eye surgery since, but at the time, I couldn’t. So then I wanted to be in the Royal Marines. But I quickly realized I hated being told what to do. I don’t like instructions, I don’t like that kind of rigid structure. So instead, I ended up starting my own business, which was never really the plan.
And to grow that business, I had to learn how to market. I was doing something I didn’t even know had a name at the time—but it turned out to be SEO. And that’s what made our company really big. So when I exited that company, I thought, “I’m going to teach people more about SEO. I’m going to go around, because I wanted to travel, because I used to travel as a kid. As I told you, I lived in Italy, in Australia. So I wanted to freelance, and just not worry about anything, and travel the world.
And then COVID hit, and that stopped all the traveling. With nothing better to do—and I’m not the sort who just sits and watches TV—I started focusing on my freelancing business instead. That grew too quickly, and I met Suganthan, my business partner. He suggested we start an agency together.” And I said: “I’d be happy to start an agency with you, but I really don’t want to hire people.” Because I’d hated managing staff in my last company, where we had about 25 employees.
So instead, we started building tools to automate as much of the work for us as possible. And then we realized those tools were valuable, and then we started releasing them, and Keyword Insights was just one of many. We’ve built so many internal tools, but we didn’t realize at first just how hard it would be to run a SaaS business. So, we actually have loads of internal tools—some really cool ones too. We just haven’t released them publicly because we haven’t had the time to build them out properly.
And that’s really how it all unfolded—one accident after another. Now I’ve got an agency, I live in London because I’m not traveling anymore, because the company is too big. And I’ve got a software tool that grew almost by mistake. That’s how it all happened.
Gianluca Fiorelli: Well, I don’t think it grew by mistake. I think it grew because it’s good—and because it filled a clear need. The tool was really doing that, while other tools came after and weren’t as strong. And I think it arrived at a very good moment—that’s why everybody adopted it.
Andy Chadwick: Yes. I appreciate that.
Gianluca Fiorelli: Okay. Thank you, Andy. It was a wonderful pleasure to spend this hour with you.
Andy Chadwick: Thank you, Gianluca. I appreciate it.
Gianluca Fiorelli: And I hope to see you in real life. I think we are maybe going to see each other at one conference or another in the near future. Let’s promise to meet up again—maybe for a new episode of this session.
Andy Chadwick: Yeah, appreciate that. Thanks for having me, Gianluca.
Gianluca Fiorelli: And thank you to all of you as well. Remember to subscribe to the channel, give this episode a like—because I think it was a really cool one—and ring the bell so you’ll be notified whenever a new episode of The Search Session is published. Take care, and see you soon.
Podcast Host
Gianluca Fiorelli
With almost 20 years of experience in web marketing, Gianluca Fiorelli is a Strategic and International SEO Consultant who helps businesses improve their visibility and performance on organic search. Gianluca collaborated with clients from various industries and regions, such as Glassdoor, Idealista, Rastreator.com, Outsystems, Chess.com, SIXT Ride, Vegetables by Bayer, Visit California, Gamepix, James Edition and many others.
A very active member of the SEO community, Gianluca daily shares his insights and best practices on SEO, content, Search marketing strategy and the evolution of Search on social media channels such as X, Bluesky and LinkedIn and through the blog on his website: IloveSEO.net.
stay in the loop