From “It Depends” to Action: Rethinking SEO in the Age of AI | Chris Green
Welcome back to The Search Session. I’m Gianluca Fiorelli, and today I chat with Chris Green, Technical Director, Senior Consultant at Torque.
Chris Green shares why he’s excited—not anxious—about AI’s impact on search. For him, it’s a return to SEO’s experimental roots: exploring new systems with fresh curiosity. While tactics are shifting, he believes the core principles—understanding users, building authority, and mapping the full journey—still hold strong.
From agency insights to smarter workflows, Chris explains how ChatGPT hasn’t made him a better coder, but a sharper problem-solver.
He unpacks why “'It depends” isn’t always the best answer —and why offering context, asking better questions, or simply saying “I don’t know yet” can be far more valuable in building trust and delivering meaningful insights.
He also shares what he looks for in a site audit and how his academic background in film theory and audience research unexpectedly laid the foundation for his SEO approach today.
Tune in for a thoughtful, engaging conversation packed with insights for navigating SEO’s evolving landscape—enjoy the episode!
Transcript
Gianluca Fiorelli: Hi, and welcome to a new episode of The Search Session.
Meet Our Guest: Chris Green
Gianluca Fiorelli: Today, we’ve got another wonderful guest. He’s British, originally from Essex, and he recently became the Technical Director at Torque.
He’s also quite well known for his SEO videos—he was one of the first to bring that classic TikTok style into the SEO world—and honestly, he does it brilliantly. He’s also a true gentleman.
His name is Chris Green. Hi Chris, how are you doing?
Chris Green: Hi! Yeah, great to be here—and wow, what an intro. Thank you.
Gianluca Fiorelli: All well-deserved! So… how’s AIOs been treating you lately? I don’t even want to say “these past few weeks”—it’s really been months of big changes.
Chris Green: I’d say once I got over the fear part… If you’d asked me about eight months ago, when everyone was talking about SGEs and what they might mean, my answer would’ve probably come from a more fear-based place.
But now? After diving into the research and wrapping my head around things, that existential dread has mostly gone. It’s been replaced with something closer to excitement—because it’s a real opportunity to explore new systems again.
It actually reminds me of when SEO version one was brand new—when we were all still figuring it out as we went. And I’m really enjoying being back in that kind of space.
There’s still a lot of work to do, and a lot of questions coming in that I don’t have perfect answers for yet—but we’ll get there.
Gianluca Fiorelli: Yes, I feel the same way. Back in the day, we had more than one search engine to pay attention to. Now, it's not just search engines—it’s much more than that. But it’s also nice to feel like a student again.
Of course, it’s challenging to stay up to date with everything. The pace of change is so fast now, much faster than it used to be. That’s probably the biggest difference compared to the old days, when things moved more slowly and steadily.
When SGE was just a test, there was a lot of fear in the air. But now, you can see it—even from publications, from people like you, and from other peers in the industry—there’s a shift happening. We’re starting to see not just “what is this” or “what does that mean,” but also “how do you do this,” “how can we work with it,” and so on.
There are more guides, practical tips, actionable insights, and even early checklists being shared. The only thing that still seems a bit behind, maybe, is the tooling—but even that is starting to catch up to meet the SEO industry's need for more reliable data in this new context.
We’re also hosted by Advanced Web Ranking—one of the oldest tools in the industry for rank tracking and analysis.
And now, Advanced Web Ranking even lets you track visibility in ChatGPT and other platforms. The tool is similar to Waikai somehow because it’s about topics more than prompts and searches. Which makes it a great example of how SEO tools are starting to evolve along with these new environments.
What Still Holds True in SEO?
Gianluca Fiorelli: You also mentioned in one of your recent LinkedIn posts that when everything is changing so quickly, it’s completely normal to feel scared. And I think that’s still the general vibe for a lot of people—not just SEOs, but even business owners and stakeholders.
When you’re facing major change like this, it’s only natural to look for the things that still hold true—the things that persist and remain valid despite the shifts. What are those core things for you?
Chris Green: Yes, so that post I made—I think it came from years of doing SEO 101 training for all kinds of people. One thing I’ve always tried to hammer home is that the core principles of SEO—assuming you’re not black-hat oriented—haven’t really changed.
Sure, the details and tactics evolve, and honestly, now I have to eat my words a little because we are in a time of significant change. But even then, if you want to steady yourself against the feeling of being swept away or overwhelmed, there are still core concepts you can hold on to.
When I look at content recommendations or predictions around things like query fan-out or other techniques for content creation, a lot of it is still pointing in the direction of what we should have been doing all along.
Things like building topical clusters—content that shows both users and Google that you’ve got deep expertise in a specific area. Content that answers not just one question but anticipates the broader journey—supporting users not just at the moment of conversion or action, but throughout the intent and maybe the next question
In theory, a lot of those things are what we really should have been doing for the last couple of years anyway. I mean, I still remember back to one of the earliest times I heard Jono talking at a conference. He was saying, you know, if someone has a dead pixel in their TV, they may not know it yet, but they need to buy a new TV.
So if I'm trying to sell TVs, surely I want to speak to the person who's Googling what a dead pixel is, or why the screen is looking strange. Like, there's that whole kind of principle of trying to own—or try to have—content throughout your customer's journey.
It shouldn’t have been something that we waited for AI Mode to start doing. However, I would say this is a fantastic opportunity to kind of revisit a lot of that.
But even so, just having a deep understanding of customer needs, the customer's journey, signs—what kinds of queries can we use to infer their level of knowledge, their level of intent? How can we qualify a customer based on that data? I guess a lot of that's still the same on the content side.
And then if we move to technical, for example, a lot of the core technical we need to worry about is, again, broadly similar. We're in the business of—we have a bot visiting our site, looking at our content, and our job is to reduce as much friction as we can. So, what's gonna stop that bot from seeing the content that we want to be recognized for?
Now, on the technical side, there are some elements that are different. I mean, broadly, LLM crawlers are just not as good as Google. And we have to accommodate for that—you know, like emphasis on not using client-side scripts to render content.
But that's been a technical SEO staple for a while. And ideologically, I don't think that optimizing for one crawler is a great tactic. I think, where budget and time permit, you should optimize for the worst crawler that hits your website. And, you know, you are always going to, effectively, be in a technically good place—or you're going to reduce a lot of overhead for those other crawlers that are, I guess, technically better. So I think there are a lot of constants around that.
I mean, we can start going into some of the minutiae—you've talked about tools. I think tools—tracking, reporting, success—are some of the ones that maybe should be the least comfortable for people right now. Because I don't think we have fully formed alternatives.
That said, everyone’s saying rank tracking is gonna struggle if it doesn’t evolve quickly. And I think, actually, rank tracking has been struggling for a while. You know, trying to get users to move away from just looking at organic position, trying to get users to move away from the 10 blue links—think about pixel depth, think about click-through rate—those are far more useful metrics.
We haven’t moved in that direction yet. And I do think that conventional method will be gone probably sooner rather than later now—assuming that AI Mode will roll out and become more ubiquitous, in whatever form it then takes.
So I’d say there are some constants—and then there’s a less constant one for you there.
Gianluca Fiorelli: Yeah, and I think you're right. Because, I mean, the classic tracking tools that just say “you’re in position one, you’re in position two,” etc.—those kinds of features aren’t so helpful anymore.
The more advanced tools, yes, they do show you things like pixel position, and I think that’s actually super interesting. Because then you can relate it to the CTR. But also, many tracking tools—well, some, not all—are just telling you where you are in the classic organic results. Some others might show you if you appear in “People Also Ask,” but that’s about it.
And I think that’s a limitation. Because honestly, clicks—and ranking as a metric—were already kind of “dead” a long time ago. But now, with AI Overviews especially, CTR is shrinking even more. So maybe now, it’s more about visibility.
But then the question becomes: how do we measure the value of visibility?
At some point, I started to think: “Okay, visibility for the sake of visibility—it’s just noise.” Maybe instead we should be asking: how much are we visible? For example, in a single SERP, we might have a presence in the AI Overview, appear in the classic organic results, and maybe also show up with an image. So with that in mind, pixel calculation could be really useful.
So, knowing how much space a feature is occupying in the SERP, we can start to say, “Okay, considering an average SERP height in pixels, we’re occupying 40% of that SERP.” And that could be a starting point for reasoning about the impact of visibility—and the degree of attention we’re capturing from people who are searching.
This might actually be the biggest shift—not just everything else you mentioned, which are things SEOs should have been doing for years now, both technically and in terms of site architecture and content. But the metrics—for me, that’s the real big change. That’s where we’re maybe struggling the most during this transitional period.
But that’s our view—from the SEO side.
Agency Perspective: What Clients Are Asking Now
Gianluca Fiorelli: What about your perspective from working inside an agency? From the business side, from the clients—whether they’re approaching you now or already working with you—how are they feeling and responding to these changes?
Chris Green: Oh yeah. Right now, I think there's still a huge degree of: “We've heard stuff about this—about AI in search—and we know AI is really big… so what does that mean for us, and where do we fit in?”
Interestingly, we’ve had more client feedback when people have actually started using AI tools themselves. You know, asking ChatGPT to generate a marketing strategy or suggest some target keywords—or even making commercial buying decisions based on ChatGPT data and things like that. I think that’s when the concern—or at least the acknowledgement that this is potentially really disruptive—has started to really land.
The big announcement about AI Mode being officially launched in the US, at least for our clients, definitely marked a shift. That made headlines, and I think people suddenly thought, “Wait, something big is changing—what does this mean?”
So, that’s generally the question right now: “What does this mean for us? How is it changing?”
Until very recently, it was tricky to really judge the impact. I’d say for businesses that were more present in AI Overviews, they started seeing that impact a bit earlier. But for some of my clients—who are less top-of-funnel and have classically been targeting more bottom-of-funnel—it’s been harder to spot disruption from AI-based search.
You know, we've looked in Google Analytics, we've looked at which AI chatbots might be referring traffic. We’ve been watching for what kind of click loss might come from AI Overviews, and honestly, it wasn’t really until around March this year—when we saw that much wider expansion of AIOs—that we really started to see meaningful changes.
Now, I’m seeing this pattern across other industries too, and from people working with similar clients: clicks are starting to drop, but impressions are starting to rise. That kind of paradigm is starting to play out. And I think people are getting generally more concerned now.
It becomes, “Okay, if we can’t track these things as well, do we even know how visible we are?” And then we circle back to the tools question—the tools and the reporting haven’t fully caught up with what’s actually happening right now. There’s still a lot of the unknown.
I’ve done some work around pulling ChatGPT data, and some of our tools can pull general AI Overview data. But for the most part, the positioning today is about steadying people—saying, “Look, it’s not having a dramatic impact on traffic right now. It will, and in some cases, it already is. But for many, it’s still emerging.”
What most people are asking is, “What are we going to do about it?”
And even though, like I said earlier, there’s a lot of continuity in theory or certainly in principle, the tactics are starting to shift. That’s where there’s a lot of interest now. It’s, “Okay, you’ve explained the theory—but how do we actually do this?”
We’re starting to see more examples from the community—people who are actively putting this into practice and sharing their stories—but I still think a lot of that is very much in the process of being figured out.
I'm certainly using this as an opportunity to say: AI-generated content, AI search results—they're coming from content. Google and other services have gone out, reduced your content into chunks, and they'll return the most relevant chunks to match a user’s query at that time.
Quite simply: if you’re not providing that information, somebody else will. Or worse, that information might just get made up—or extrapolated—from another source entirely.
So there’s a degree of: I’m using this to really hammer home the point that—if visibility is now key, and we can’t rely on clicks as a solid measure—then you have to show up. You’ve got to put that content out there.
And it's not just about long-form—it's not just, “write 2,000 words on X,” because SEOs have had that in their playbook for years. It has to be smarter than that.
Like, do we really understand what pieces of content, what snippets, we need across the user journey? And we also have to factor in personalization a bit more.
So yeah, we’ve definitely got enough to start having these conversations with clients. And honestly, I think we’re in a really good position. Out of everyone in the mix, SEOs are the best positioned to answer these questions.
Google doesn’t even have all the answers yet. So I think everyone’s lagging a little behind the speed of development. We’re all kind of... catching up.
Gianluca Fiorelli: Yeah, and I think, at least for me, one of the nicest realizations is this: yes, we’ve built a strong technical foundation, but now—we’re doing marketing again, finally.
Because if visibility is the goal—and visibility just for the sake of it is, like I said earlier, just noise—then you need to nudge attention. You need to be memorable.
So now, even in terms of metrics, it’s: “How much is your direct traffic growing? Is there a correlation between that and your visibility in search? Are branded searches increasing?”
That means people remember you, and hopefully remember you positively.
And that ties into what you were saying earlier about personalization. Because Google, ChatGPT, and other systems rely heavily on memory. So, being part of a user's memory—even at the browser level—that’s key for SEO now.
We might not even need to rank... because we’ll already be suggested. “Hey, this brand also offers that,” or “This person talks about this too.”
And that’s marketing—true marketing. Because if you want to be remembered, you can’t be like a hundred other brands. You need to have your own voice. You need a unique selling point. And maybe it’s not even your content—maybe it’s your customer care, or how you handle user experience on your site.
Of course, that still includes usability and technical excellence.
How Chris is Using AI to Build Workflows, Not Just Content
Gianluca Fiorelli: But now, shifting gears—let’s talk about AI again. How are you personally incorporating it into your daily work?
Because I think there are two kinds of SEOs here. There are the manitas—as we say in Spanish—the technically gifted folks who jump into APIs, experiment with models, use Python, and build their own tools. That’s not really me—I’m not great at that stuff—but I do have my own vibe, my own way of using AI.
What about you?
Chris Green: Yeah, so for me, the biggest shift has been using AI to build workflows—from small ones to more complex setups. There are a lot of people doing interesting things in that space right now.
I’ve got some background in Python, JavaScript, command line tools, that kind of thing—but I’m not a gifted engineer or programmer, and honestly, I never will be.
So for me, the real opportunity has been in finally being able to bring to life a lot of the ideas I already had. I already knew what I could pull from APIs, what data was available, and how we might approach certain problems—but AI, especially assistive coding tools, have helped me actually execute on that.
Over the last six months or so, it’s not been about becoming a better programmer—it’s been about becoming a better prompter. I kind of hate that term, but it’s true. I’ve been learning what the AI assistant is good for, what I am good for, and how we collaborate and ideate effectively. And also—what are the common pitfalls going to be, and how does it build in space for error?
Because AI isn’t perfect—it can hallucinate, and its overconfidence can make it risky. So when you're calling data from an API, you need to ask yourself: “Am I saving every API call that’s made? Am I keeping all the raw data to review it?
If you’re running an analysis, ask: “Have I broken down each stage of the analysis so you can kind of really understand when you need to go into the debugging work? Why is it not working? This AI assistant isn't telling—you can kind of do that yourself.
So that kind of thing, I really feel like I've built up a lot of, and it's been quite liberating, to say the least, in terms of what it's starting to enable me to do. There are a lot of challenges that I'm solving, but the key is, it's small things.
Like, you know, in particular instances, I just need to crawl sitemaps really quickly. I know I had tools that could do it, but for this specific workflow and the particular things I needed to analyze—and I had to repeat it 10, 20, 30, 40 times—it was just easier to write something in Terminal that could run it.
That worked nicely. Other people said they needed it too, so I got ChatGPT to convert it into Streamlit—into Python that worked with the Streamlit app—then hosted it on Streamlit so I could give it to other people.
That kind of thing has really sped things up. And I’ve said this to a few people now: it won’t be long before agencies, consultancies, even some in-house teams—we’re all becoming more like workflow builders, rather than something else.
And I don’t say “just workflow builders” like it’s something minor. We all have the ability now to do far more powerful things with SEO, with tools, with data sources. We can cover more ground far quicker, or say more insightful things without the cost or time overhead.
But we’re all in that same ballpark now. In theory, anybody can kind of do it—it’s democratized to an extent, that level of working. But now the question becomes: if anyone can build a workflow, where's the value? And it’s more about how we tailor these workflows to the business or commercial context they exist in.
You know, I can post a video about how I’ve done redirect mapping for hundreds of websites using vector embeddings—kind of finding the best matches based on that. But the process I’ve built for my clients is highly specific: specific to their URL structure, to their business, and website logic. It has exceptions, it has rules. I’ve used vector embeddings for some parts of the matches, and I’ve used fuzzy matching for others, because there are certain scenarios where I want one over the other.
That really specific technical detail? It’s not useful to anybody else outside that exact situation. But I’ve been able to do that now because of what these assistive AI tools have enabled me to do.
So there are other smaller things too, but that’s how I’ve been using it. Just opening your mind to, “I couldn’t solve this problem before”—but now, maybe you can solve it in a completely different way. So, revisit those old problems and see what you can do differently.
Gianluca Fiorelli: Correct, correct. That’s pretty much the same attitude I’ve developed toward using AI. Over time, I’ve realized that more than just describing the task well, the most important thing when prompting is being able to define the context clearly.
Now, I’ve started treating Gemini, ChatGPT, or even Claude as if they were my own very junior SEOs. I explain everything to them in great detail so they can do the job, and then I review their work afterward.
And I think a really useful exercise, just like we used to do back in the day when we tested things on our own websites and playgrounds, is to take that same mindset but apply it to AI tools. For instance, take a topic you know really well—so you won’t fall for hallucinations—and use that to create different types of prompts. Especially when it comes to outlining or updating site architectures.
That way, you can judge whether what ChatGPT suggests actually makes sense. And through that process, you also start to understand how the machine works—why it chooses one thing over another. You’re not just getting the output, you’re learning the mechanics behind how these LLMs function. And that’s knowledge you can later apply directly to client work on those topics.
So, let’s think about how we can optimize and work with chunks—the classic idea of breaking content down. I actually prefer the term “snackable content,” because LLMs are like Pac-Man, just gobbling up everything we feed them.
Doing so, it’s become clear that one of the most valuable things we’re gaining is time. Yes, we might spend a lot of time upfront figuring out the right things to give to AI—what prompts to use, how to structure input—but once we dial that in and start getting useful output, it’s incredibly freeing. It gives us more space to focus on where the real value of our work lies: offering our own insights.
It’s like when crawlers like Sitebulb or Screaming Frog came out—suddenly, we didn’t have to rely on just Xenu and figure everything out from scratch. These tools did the heavy lifting so we could spend our time interpreting the data.
And with AI, it's the same: if it gives us back some of that time, we can use it to be more insightful, more creative—even more strategic. That’s probably where our best outcomes as SEOs will come from.
Chris Green: Yes.
Chris's Approach When Reviewing a New Client's Site
Gianluca Fiorelli: Let’s switch gears and move away from AI for a moment. I’ve been asking this to other guests recently: a new client comes in—you’ve done the briefing, you’ve got everything—so this time you're seeing the site with fresh SEO eyes. What’s the first thing you check?
Chris Green: Oof. Well… if I have access, Google Search Console is up there in the first few things—not the first, but definitely a priority.
But first, I usually dive into the homepage, click around, and test interactions. I often disable JavaScript, then re-enable it—that’s one of those old-school habits. I also check the robots.txt file and poke around key pages to get a sense for the site.
That’s where my technical background is. I try to get a feel for what the website is. Can I figure out, just from the site and the experience on key pages, what I'm supposed to do and where the website thinks I am in the user journey?
Because historically, when I was more agency-focused than consultancy-based, a lot of my clients were B2B—very niche, very specific businesses. And I just got so used to seeing websites built under the assumption that everyone already understood their jargon—the business’s internal language. They assumed users arrived knowing exactly what they needed.
That’s changed a bit now; I think the overall level of maturity has improved. But even today, I still find that’s where a lot of websites fall short. Most technical issues are solvable—or at least not so significant that they block progress entirely. I mean, we’ve all seen poorly built sites rank really well, and technically perfect ones that rank for nothing.
So the core of where I spend most of my time now is understanding: do we really know how users are finding us? What’s their level of knowledge? What are they trying to achieve?
But the very first things I still can’t help doing—like checking how much JavaScript the site uses, what paths or directories are blocking the robots.. There’s still a lot you can learn from that. Though I admit, I don’t do it quite as obsessively as I used to.
I think, like many junior SEOs, when you first learn about robots.txt, sitemaps, meta tags, and all that, you can't help yourself—you dig in and chuckle when you find a site still stuffing the keywords meta tag. But these days, my focus has shifted more to: does this site even know who its customers are? And you can usually get a sense of that pretty quickly just by looking.
Doing SEO in a Fragmented Search Ecosystem
Gianluca Fiorelli: Another shift I’ve noticed is that when we talk about organic search, we no longer think just about classic SERPs or rich results. We’ve expanded to include video—think YouTube appearing in SERPs, and so on.
But now, with the newer generation, people are searching in other environments: TikTok and Instagram are being used to find things to do, eat, and places to go. So I wonder—how much are you still focused on that classic framework—your website and Google search—and how much have you started expanding your analysis and insights to include other platforms? After all, these non‑Google environments are part of the modern search journey.
Chris Green: Yeah, that’s an interesting one. Right now, a lot of my clients are still fairly siloed in terms of their digital teams—social and paid are often handled separately. We have the most flexibility and ability to impact in the channel or the remit that we've kind of been brought under, which adds a point of clarity, but kind of frustrates things.
Like you said, traffic from traditional search—what we used to think of as 10 blue links—is dropping, and that’s largely because the search landscape has fragmented. One area I try to get involved in early is feed-based features—things like Google Merchant feeds, hotel listings, job schemas, and of course, Google Business Profiles. I still find myself pretty grounded in the Google ecosystem, but that’s shifting quickly now toward chatbots and other AI-driven interfaces.
One simple but important piece of advice I give: fill out every data field Google offers—product details, business info, structured data, all of it—with the most accurate, reflective information you can. It may sound basic, but we’re in a world where Google and others have so many sources to pull from. If you don’t provide it yourself, they’ll find it elsewhere—and it may not reflect you accurately.
So, I think pushing into other areas is something I definitely do. I tend to be more on the technical and data side, so I’m often building insights from other platforms—trending hashtags, trend data, merchant data, and ads data.
I try to get closer to BigQuery or whatever database setup the client has, because even if we’re not working on those sources directly, they still impact us. SERP is a crowded space now. There are so many features that aren’t traditionally organic.
And marketing activity from other media—paid, social, etc.—can really shift trends, interest, and demand. You can’t be truly effective unless you’re at least monitoring and reporting on that.
A very common case is branded paid activity: the client increases it, but most of their organic traffic is branded. Suddenly, organic drops. Why? Well, it's likely because those new paid ads are pushing down the organic listings, so you lose clicks and end up paying for what used to come for free.
That kind of conversation is easy to have, even if it's not so easy to fix. Another area I focus on is Merchant Feeds. Google Merchant Center gives you a price differential: are your prices generally higher or lower than competitors targeting the same inventory? And when that data shows up in shopping results or gets pulled into AI-generated answers, it's going to affect clicks.
People are going to click on cheaper versions of the same product—if given that option—there are obviously trust, ratings, shipping, and other factors. But price and availability remain really key. So if organic traffic, revenue, or conversion rate is dropping, I need to ask: are we price‑competitive? Do we have stock in? Are our delivery and returns policies in line with others?
I don’t want to use the word “holistic” because I still work in a fairly narrow niche, but there’s a lot of external data we can pull in—even if it’s just to help guide our own channel strategy.
I still believe the sweet spot is looking at as much of the user journey as possible. The channels we work with—paid search, paid social, organic search, organic social, email—those are just swim lanes we've set up based on how we've organized our teams. But users don’t see those distinctions—and they shouldn’t.
So there’s a strong case to be made: if we truly want to be user‑centric in how we market, we need to follow users wherever they go, not just wait for them in our preferred lane. There's still a degree of that, and a lot of organizations are built that way. So, redefining that structure in large organizations, however, is hard, and it takes years.
Reflections on Chris's “It Depends” Short-Form SEO Video Series
Gianluca Fiorelli: Yeah. Yes, I completely agree. I want to ask you something, since we were talking about different content formats. You’ve experimented with short-form video—those vertical TikTok-style clips—for your personal brand. And I have to say, the name of your series, “It Depends,” was brilliant. So, I’ve got two questions: First, what kind of impact did that have on you, especially putting SEO content into non-traditional spaces like that? And second, what’s the kind of “It depends” question that really makes you roll your eyes—like a gyroscope?
Chris Green: Great questions! First off, the impact of doing the videos: SEO is a pretty small space, and I’m definitely not chasing clicks or trying to go viral. I have no illusions—some bearded guy offering slightly thoughtful takes isn’t exactly TikTok material.
But anecdotally, I’ve met quite a few people who say they’ve seen the videos and got something out of them, which is great to hear. I think the intent behind “It Depends” was really about answering those classic grey-area SEO questions. Hopefully, it helps people understand those topics just a bit better.
I’ve benefited greatly, personally, because for every 60-second “It Depends” video I make, I have to really actually understand enough of the angles to explain it succinctly. I very quickly realized that was what I took from it—forming your ideas into a 60-second chunk that loops. It’s a slight gimmick of the format, but it’s quite a nice degree of mental agility.
Whenever someone throws an “It depends” question at me—especially one that’s hard to answer without using those exact words—I hang onto them, I save them. I’ve actually got a running list of them saved. I haven’t made a new video in a few weeks because I’ve been swamped, but that list is my go-to for the next batch of ideas.
The “It depends” question that causes me to roll my eyes? That’s a challenging one, because there are so many. I think the part about “It depends” as a phrase that makes me roll my eyes the hardest isn’t necessarily what it’s in response to, but how it’s used. Sometimes, of course, the answer is nuanced—it can be super complicated. And sometimes the question itself just isn’t answerable in a straightforward way. Like, someone wants an easy answer, and it just doesn’t exist.
Handling the Dreaded 'It Depends' Questions
Chris Green: What’s always grated on me a little bit is when “It depends” is used almost as a meme response. Sometimes, when there’s a difficult question—or someone simply doesn’t know the answer—they’ll say 'It depends' with a slightly smug, knowing tone, as if that alone is a satisfying response.
But in my experience—whether it’s stakeholders, clients, colleagues—no one really appreciates that. We might snigger a bit, especially at conferences, when the first person says “It depends”—it usually gets a chuckle from the audience. But “It depends” often masks “I don’t know the answer” or “That question’s hard to answer.”
That’s always grated on me, because it’s actually an opportunity. You could say, “Tell me more about the situation,” or, “Can you explain the context?” It forces SEOs to do something they often find hard: have an opinion and own the answer.
If someone says, “It depends,” I want to ask: What do you know about the person asking the question? What background or context do you have? How can we strip away some of the “depends” factors? Let’s make some assumptions. Let’s build a scenario where you can actually provide a meaningful answer. If you know someone works client-side, or in an agency, or in a large business, or D2C—those details can help you go from “it depends” to something that’s genuinely helpful, and then let's have that conversation.
I think I’ve kind of dodged the question a little bit. I guess the ones that really bug me are the classic ones, like, “How long does it take to work?”—that’s the perennial one. Or, more recently, “So once I’ve done this, when am I finished SEOing the site?” Those two are the real throwaway ones. But yeah, the more thoughtful answer is: it annoys me that we use "it depends" as a response at all.
Gianluca Fiorelli: Yes. In fact, I think everybody uses "it depends."
Chris Green: I nearly did. I nearly said it just then as well—I stopped myself.
Gianluca Fiorelli: What I’ve learned, and what I’ve seen works better as an answer, is to lay out the options. Like, give me more context, and I’ll try to answer you. Sometimes, it’s better to just be honest and say, “I don’t know.”
In this moment, I just say, “I don’t know—please, give me time and more context. And with that, I’ll be able to come back with something meaningful.” Also, in many cases—especially when it’s a topic I know a bit about, but I’m not the deepest expert—I’ll just say, “Let me inform myself.”
I know who the real experts are. If it's something like AI information retrieval, or similar, I’ll turn to people like Dan Petrovic, Michael King, or Andrea Volpini and the guys at WordLift. They know this stuff ten thousand times better than I do. So I’ll say, “Give me the time, give me the context, I’ll dig in—and then I’ll give you the best answer I can.”
That usually helps build more trust—whether it’s with a client or your boss—rather than just defaulting to “It depends” and giving a rushed answer.
Chris Green: Absolutely. The thing is, SEOs—especially the more technical ones—are generally super thoughtful. We want to give answers with authenticity and substance. But saying “I don’t know” can feel scary.
If you’re seen as the subject expert, admitting you don’t know something can feel like a risk. But honestly, you’ve got to be confident in your role to say it. And I 100% agree—it’s about saying, “I need more time for these reasons, but I will come back with an answer.” That’s actually empowering when you get to that point.
When Best Practice Isn't the Best Choice
Gianluca Fiorelli: And talking about the classic frame of mind, this is just the last question. We’re always used to thinking in terms of best practices. But when reality shows a different situation—especially on the technical side—sometimes best practice isn’t the right move for the job at that moment.
So, let’s be the uncles—or in my case, grandpas—of SEO, talking to the younger generation. I mean, best practices are paramount and we should know them, but how would you suggest they also think tangentially or laterally? How do they know when best practice isn’t the best practice? How would you suggest these kinds of things—especially to the younger generation in SEO?
Chris Green: So when I train really new SEOs, especially around audits, you know, the hardest part for them isn’t finding what’s wrong or right—it’s how to understand severity and urgency. What’s the real priority?
You mentioned Sitebulb and Screaming Frog earlier. They’re fantastic—they crawl the site like Google, highlight issues, and even try to approximate severity themselves. But it takes time to learn, but I always encourage SEOs to think: “What's the context here? What's the business context behind that technical issue?” Some things are really easy to say whether they’re right or wrong—like, you're missing your XML sitemap. That’s wrong. It's a clear-cut issue: you don’t have one, and it’s recommended that you do. Okay, fine.
However, then you have to ask: “How long is it going to take that business to create an XML sitemap? How easily can they maintain it? And if we do have that sitemap, what does it actually change?”
Once you start asking those kinds of questions, it exposes the need to understand the purpose behind each of these things. And that takes time, because there’s a lot of foundational knowledge to absorb.
Take the sitemap example again. Its main job is discovery—it helps Google find the pages on your site. So the next step is to ask yourself: “Is the missing sitemap causing pages to not be discovered at all, or just not being discovered as quickly as they could be?”
Then you go down the rabbit hole: is the navigation crawlable? Are there orphaned pages—pages that aren’t linked from anywhere and can’t be found from a crawl starting at the homepage?
You start to realize there are scenarios where a missing sitemap might be technically non-compliant or not best practice, but it might not have a clear or measurable impact on performance.
And I learned that the hard way. My first job in an agency was doing website audits. Within my first two weeks, I was given a technical audit, handed Screaming Frog and Xenu, and I went through it, found loads of issues, wrote them all up, and sent them straight to the dev agency that built the site—not to the client.
And it went really badly. I thought I’d absolutely knocked it out of the park—I’d found loads of issues. But I did a really poor job at prioritizing them. A lot of the issues? The site was built in .NET, so fixing them would’ve taken forever and yielded no real benefit. And I was also really, really annoyed the developer in the process.
I didn’t have the business context to make the right recommendations, and I communicated it all in a way that was never going to get the response I wanted.
So I think it’s really about focusing first on why something is a best practice. Do you actually know what impact it has? If you don’t, then the more time you spend learning that, the more incisive your recommendations will be later—and the better you’ll be at prioritizing.
And it just takes time. But that’s the key thing. If anyone’s looking to get into technical SEO, that’s what they need to work on: building that foundational knowledge.
Gianluca Fiorelli: Yeah, yeah. Perfect. And I was just picturing the devs in that case—rolling their eyes.
Chris Green: They did more than that! I even had one or two cross‑phone calls to deal with at the time—that was fun.
The Fireside Questionnaire
Gianluca Fiorelli: Okay, now let’s go a bit personal. For our listeners and viewers, we actually share quite a few things from our university days. You told me a few days ago that you studied semiotics for resistance—just like me—but I also discovered you studied the history of film theory. I did too: my specialty was the history of cinema, and I wrote my thesis on it.
Chris Green: Oh, don't test me on my history.
Gianluca Fiorelli: No, no, but my question is: which of your academic studies suddenly feels relevant to your work now? I know for me that linguistics and multimedia knowledge—pretty much dormant at first—became valuable only in recent years. But now, all of a sudden, all these things that I study are becoming relevant. Has the same happened to you?
Chris Green: For me, a lot of what I studied was around meaning-making—the idea that things can have multiple meanings depending on the viewer. That’s been incredibly useful for critical thinking.
When you start looking into customer journeys, understanding user intent, or getting empathetic with a user at a specific point, that perspective becomes invaluable. I think approaching text, webpages, film, or whatever with the mindset that “I might take one meaning from this, but that meaning isn’t universal—not even the creator’s intended meaning”—is incredibly powerful.
Actually, the biggest insight came from my film studies, where I did ethnography and audience research. I looked at adaptation theory—how books become films, video games become films, films become books. Like any kind of adaptation, how one medium is moved to another.
But the audience research part I did—well, during my undergrad, I didn’t even realize it at the time, but I was effectively scraping web forums to monitor sentiment around adaptations.
I didn’t know what SEO was back then. But that ability to scrape content, measure sentiment, and mine forums for insights into what people liked or didn’t like—that ended up being really formative for me, even though I didn’t realize it at the time.
When I did my postgrad, my master’s, it was more focused on qualitative research—specific audience questionnaires, one-on-one interviews. And again, a similar theme: How do you design questionnaires that don’t lead people? How do you understand what people actually think without introducing bias?
That’s helped me with analysis and content creation. I used to do more link building through content, and these research skills really fed into that. And all of that still underpins what I spend my time doing now.
And that’s before we even get to things like reading academic texts—referencing systems, how to do research. You can put all that aside, but those core skills—broadly, anyone with a humanities degree might have picked them up. And yeah, we can all learn to code or build websites. But those foundational skills—understanding audiences, uncovering insights, analyzing data—I use those all the time now. I 100% see that as incredibly valuable.
Gianluca Fiorelli: And when you’re not working on SEO, or thinking or writing about SEO, what do you like to do?
Chris Green: Oh—well, I’ve done that thing that people seem to do in their mid to late thirties, where you suddenly feel like, “Okay, I need to take ownership of my health.” So I’m much more active now—exercising five or six times a week.
Besides that, I have two kids who take up a lot of my time. I still play video games every now and then—not nearly as much as I used to, but I get in a bit when I can.
Yeah, those are the big ones. And honestly, when you ask what I do when I’m not thinking about SEO—lately, it’s been sleeping. It’s been pretty intense. I tend to get really immersed in things—it’s not just a 9 to 5 for me. My brain keeps ticking on it even outside of work hours.
Gianluca Fiorelli: Yes, and I mean, when doctors say people should consistently get at least seven hours of sleep a night—they’re right. Because if you don’t, you risk burning out. Even if you're technically not working, the exhaustion alone can wear you down.
Speaking of games, just out of curiosity—what’s your all-time favorite?
Chris Green: Oh wow, that might be the hardest question you've asked! I think there are two that were the most formative for me: one is Half-Life on PC, and the other is the original Halo. Those were big ones.
But honestly, we could go on for half an hour just listing games. I know Jono and Mark have asked similar questions and we’ve been on for hours. You’d definitely see evidence of my misspent youth there.
Gianluca Fiorelli: Yeah, yeah. That could be a special The Search Session—the video game edition.
Chris Green: I think there are a lot of SEOs who would jump at that. I mean, gamers have infiltrated everywhere.
Gianluca Fiorelli: And just one last question about you. What’s the sound you love most? It could be a natural sound, a song, something that reminds you of something special. What is it?
Chris Green: Hmm, this is kind of clichéd, but I’d say morning birdsong. I leave the house around five or a quarter past five every morning, and especially in the summer, the sun is just starting to rise, and the birds are just starting to sing. That’s probably the calmest, quietest part of my day. I have to wake up early to get it, but whatever mood I’m in when I wake up, within 15 minutes of being outside, that sound always centers me for the day. There’s just something about it.
The other one would be the sound of running water—rivers, the sea. That’s probably my second favorite. But I don’t live near running water, so I’d have to travel for that. So I’ll take the one that’s on my doorstep: morning birdsong.
Gianluca Fiorelli: Okay. Thank you, Chris. It’s been wonderful to have you on this episode. I think you gave us—and our listeners and viewers—so many thoughtful insights and plenty to reflect on. It was a truly pleasant conversation. Thank you, Chris.
Chris Green: Thank you for having me. Cheers.
Gianluca Fiorelli: And thank you to all of you for watching me—and Chris. Don’t forget to subscribe, click the bell, give the video a like, and I’ll see you in the next episode.
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