AI SEO

[00:00:02] Nell:
OK. Hi, everyone, and welcome to today’s webinar, Navigating SEO’s Evolution in the AI Era. My name is Nell Lindquist. I’m the Director of Marketing Operations at Liquid Web, and I’m excited to be moderating this panel today. A little background on Liquid Web: we’re a premium hosting provider for businesses of all sizes, offering reliable solutions like dedicated servers, cloud hosting, and VPS hosting, all backed 24/7 by our knowledgeable, friendly, and helpful support team.

A bit of housekeeping before we jump in: you’ll see the chat section on the right side of your screen. Feel free to use that throughout the discussion today—it’s visible to our presenters and everyone in attendance. You’ll also notice a Q&A section on the side. We’ll use the last 10 minutes to answer as many questions as possible. If you see a question you like and want to ensure it gets noticed, you can go ahead and upvote it.

Today, we’re going to explore how AI is transforming SEO and what businesses and SEO professionals can do to thrive and adapt in this new landscape. Our panelists—Lindsay, Nate, Ryan, and our very own Amanda—will guide us through the discussion. Let’s start by going around and having each of you introduce yourselves and tell us a little bit about your work. Lindsay, do you want to kick us off?

[00:01:28] Lindsay:
Sure. Thanks so much, Nell. I’m Lindsay Halsey with Pathfinder SEO. I focus on SEO for small and medium-sized businesses in the agency context, as well as working with web designers and agencies who want to offer SEO services.

[00:01:44] Nell:
Awesome. Thank you. Ryan?

[00:01:46] Ryan:
Hey, what’s up, everybody? I’m Ryan Jones, based in Detroit. I lead the SEO team at Razorfish. I also do some SEO tools and whatnot in the marketplace that I’d be glad to talk about later.

[00:01:58] Nell:
Great! Thanks, Amanda.

[00:02:02] Amanda:
I’m Amanda Va Global, Director of SEO at Liquid Web. I’m helping Liquid Web overtake the organic search space for hosting. I’ve been doing this for a little over 20 years, and I’m really excited to be here with everyone.

[00:02:18] Nell:
Thanks for joining us, Nate.

[00:02:20] Nate:
Hey, I’m Nate Day, founder of Syllabi and Profound Strategy. Our new software platform is focused on SEO content teams, small to large, looking to create the best content in the world. Engagement and having the most helpful content seem to be key ranking factors today, and we’re building a platform to support that capability.

[00:02:46] Nell:
Wonderful. Thank you all for joining us. I’d love to start the conversation by discussing the landscape. In this new era of AI and generative AI, what are the big hurdles and opportunities you’re seeing for your profession and businesses in general? Amanda, would you like to kick us off?

[00:03:08] Amanda:
Absolutely. I feel like AI is impacting SEO by transforming the strategies we’ve relied on for years. At Liquid Web, we’re using a variety of tools and prompts to help with topic research, content generation, SEO optimization, and deeper analysis of user intent—all with the goal of helping our customers make informed decisions. The foundation of SEO hasn’t changed, even with AI. It’s still about providing the best information to help our customers.

What AI isn’t, though, is a blanket replacement for SEO practitioners. One of the biggest challenges businesses face is scaling without assuming that AI can fully replace human input. Google’s March 2024 core update was clear: they don’t want spammy, low-quality content. The tools can help refine strategy, but practitioners are essential for identifying what’s unique and valuable. AI should challenge us to evolve and find ways to better meet user queries and improve decision-making processes, reinforcing trust in our brand.

Ryan, Lindsay, Nate—what do you think?

[00:05:25] Ryan:
Yeah, I’ll jump in. The biggest question I’m hearing is, “How should I be using AI?” We’ve seen people using AI for great things—like content outlines—but we’re also seeing people force AI into places it doesn’t belong. AI should be doing the repetitive, boring tasks so we can focus on creative work. But what’s happening is the opposite: AI is writing poetry, and I’m still doing laundry! Hopefully, today we can give people some tips on how to use AI effectively.

[00:06:33] Lindsay:
Building on that, SEO has always been in constant evolution, and AI is just another step in that process. For small and medium-sized business owners, it feels like even more to keep up with. There’s also a segment of people looking for shortcuts in SEO, and AI is being used to cut corners on one of the most human-intensive parts of SEO—content creation. But customers want genuine, helpful content, and AI shortcuts can be dangerous for long-term SEO strategies.

[00:07:42] Nate:
I wonder if AI has actually saved us time in SEO or if it’s made us less effective overall. If Google is looking for the most helpful, engaging content, and there are already hundreds of great pieces out there, expecting AI to create something better is a stretch. AI is not human, and it doesn’t understand the nuances of human experience.

If we step back and focus on our objectives, we can use AI to get there. But SEO can be a busy, chaotic space, and it’s easy to get lost. AI is dangerous in this sense because we think it will change everything, but it hasn’t—at least not yet. We’ll mature as a channel, but it will take time.

[00:10:10] Ryan:
Quick example: I saw a tweet today where someone asked AI for SEO advice, and it gave the opposite of what Google recommends! AI trains on bad SEO advice just as much as good advice—there’s a lot of bad SEO out there.

[00:10:33] Amanda:
Right, it’s quality over quantity. We’ve got to keep that as the North Star in SEO.

[00:10:53] Nell:
Building on that idea of quality, are you seeing or anticipating any changes in how algorithms are going to rate content because of the noise created by AI?

[00:11:06] Nate:
I think it’s just intensified what already existed. Google’s guidance on great content was already vague because there’s so much content out there. What AI has done is intensified the insanely competitive nature of content and user experience. Exceptional content and user experiences are ranking well right now, and if you’re not more engaging and helpful, you’re not going to stand out. AI is adding more content to Google’s index, and I think Google will benefit from this—more data and more training volume to identify great content. Google’s going to get even better at discarding the mediocre. We all know hanging out on page five is meaningless, even page two doesn’t make much impact. Traffic goes to the top, and with AI generating more content, that sharp peak at the top is becoming even sharper. It makes figuring out what people love and giving them that content even more challenging.

[00:12:30] Ryan:
I think it depends on the approach. The typical SEO answer, right? For those who don’t know, before I did SEO, I was a software engineer, so I can code and love geeking out about how search engines probably work. What we know is search engines don’t know if content is quality just by looking at it—they fake it with things like links, clicks, or other signals. There are a million AI detectors that claim to detect AI-written content, but none of them are good. They’re not bad because of poor coding; it’s just a difficult problem in computer science. AI picks the most probable combinations of words, and those words are used by humans too. So, distinguishing AI-written content is tough, and I don’t see Google, Bing, or any other search engine ever building a true AI detector.

The problem isn’t easy to solve. You can run AI content through Shakespeare or the Bible, and AI detectors will flag it as AI-generated, which tells you something. The issue is difficult to solve, but when Google runs its machine learning algorithms, most AI-generated content is flagged as unhelpful or low-quality. So yes, algorithms will change, but not drastically because of AI detection. The increase in data will be exponential, but we’re not going to see profound changes like detecting AI content.

[00:14:38] Nate:
Yeah, I met with a business owner this morning who signed up for a program that identified keywords, generated content, and even published it on their site—all automated. It felt like I was in an episode of Black Mirror for a moment. There are a lot of well-intentioned marketers believing the sales pitches of tools that promise to automate SEO, but it’s the same old story. If automating SEO were that easy, 1,000 competitors would be doing it too. In a zero-sum game, only a few pages win, and the easier it is, the lower the quality. It’s great if you’re on page two or the bottom of page one, but many never move past that. That’s where the real work begins. Amanda can speak to this—getting to page one is great, but that’s not where we stop. We have to move from ranking to driving traffic and business. AI has only made old SEO myths even worse.

[00:16:32] Nell:
Yeah.

[00:16:33] Amanda:
It’s all about deep analysis. You need to leverage tools to figure out what isn’t being answered and then use your intuition to survey and provide data. The goal is always to give people the information they need to make informed decisions, but in a way that’s engaging, creative, and unique. Content should feel like a book people want to return to. The machine will get better at identifying intent and user satisfaction, and over time, it will evolve quickly. As humans, we need to evolve too—using these machines to serve people better and help them make purchase decisions.

[00:17:59] Nell:
It sounds like you’re all in agreement that this isn’t going to reduce work—it’s just going to change the work, making AI a more strategic tool.

[00:18:08] Lindsay:
That’s right.

[00:18:09] Amanda:
The companies and people who hone in on this will succeed. The ones who think they can replace people with AI to run faster are going to fail—they’re going to fall off the search results. We’ve seen this before. Gamers drop off, and the people producing quality content continue to succeed.

[00:18:38] Lindsay:
We’re already seeing it.

[00:18:39] Ryan:
Go ahead.

[00:18:41] Lindsay:
We’re already seeing it with small- to medium-sized businesses. Some clients decided not to work with real copywriters and instead used ChatGPT for blog content. A few months later, the difference in performance is obvious. When you move from a copywriter who is an expert in their field to AI-generated content, it shows. We’ve been here before, with businesses thinking they can outsource content for a fraction of the cost. That was a short-term win for some, but it didn’t last. We’re already seeing clients hesitate to switch, and they’re more cautious than they were 6-12 months ago. The idea now is to use AI as a tool to do less of the mundane stuff, making the job more strategic and fun.

[00:20:13] Ryan:
Exactly. People got their expectations wrong. AI is like giving a roofer a nail gun instead of a hammer—it’s not replacing the roofer, just making the job easier and faster. But as marketers, we sometimes think backward, like AI will solve all our problems. But the real question is, is this what my users want? Is this helping them accomplish a task or make a decision? Most AI solutions I see don’t help users; they just churn out spam faster. If that’s your business model, fine, but for real companies, AI isn’t a substitute for quality content.

[00:21:10] Nell: Right. Well, let’s talk a little bit more about AI as the calculator or the nail gun. It seems like we’re in agreement that content generation is not the path forward, or not where you all are spending your AI efforts. So, where are you utilizing it? And how are you seeing it improve your workflows and outcomes?

[00:21:29] Nate: I think AI has a lot of potential. I’ll test a word out, and you all can let me know if it’s a good word. It seems like there are novel rank factors at work that separate the elements. For example, here’s one keyword, and you’re writing content for it. Should you talk about a certain subtopic? How long should you talk about that subtopic? Should you include a definition or not? Those lists of factors that are really distinguished. Should I include a video? What type of imagery should be there? How should the whole page be laid out? What order should it be in and with what style? There’s a lot of hidden, yet critical, ranking factors that make a massive difference. AI can help us identify these hard-to-find ranking factors.

I’ve found that some of the smartest SEOs I come across, who are consistently successful, study winning content. They have a process or mindset, pulling out the characteristics of winners and incorporating those into content that’s even better. That’s part of why we started Syllaby: to create a platform for that research process. I think AI is just scratching the surface here. It’s going to be a long journey, but if you can identify those novel ranking factors and make them concrete for analysis, I think AI can help us accomplish creative solutions toward our goals—not just to do a task, but to actually achieve a goal.

[00:23:27] Ryan: Yeah, someone asked a good question about custom training models for replicating brand voice, so I’ll answer that here and cover both questions. I think these types of uses are great cases for AI. AI is more than just content generation or large language models; there’s also machine learning. I believe custom models are the future—specific models designed for very precise tasks, rather than one broad model for everything.

For example, we’ve trained one on all of our brand guidelines. If you’re writing for Jeep, it will ensure that “Jeep” is used as an adjective, not a noun. Another use case is rewriting content to meet persona definitions or intent, based on parameters we provide. One more example: I’ve vectorized all my RFIs from clients and gave them to an AI. It can now access every question I’ve ever answered for a client, providing me with an outline or framework to start from. There are countless tools out there for this. I’m using Llama on my desktop, but there are public tools like Writer and others that integrate well.

[00:26:01] Nate: I love how those are well-defined use cases with specific purposes. What’s interesting is that, Ryan, those examples you gave are really helpful for non-competitive tasks. For example, adjusting brand voice, reviewing content, or iterating. These tasks are incredibly useful, but they aren’t about competing at a high level. When it comes to creating something like a novel or a piece of art—where there’s no clear good or bad output—those cases are probably not the best place to use AI. But if the output can be evaluated as clearly good or bad, AI can be very helpful.

[00:27:35] Amanda: I agree, and Ryan hit it on the head. There are so many tools out there, and some of the tools we’ve loved for years are now integrating AI, like SEMrush or Ahrefs. What I focus on is finding the tools that help you move faster and identify gaps and opportunities. That’s why I love Nate’s tool, Syllaby. It aggregates the search landscape for your intent and visualizes it, helping you move faster in identifying gaps that content isn’t addressing. It marries qualitative and quantitative research, showing you exactly what needs attention to make a page perform better.

Another tool I’m excited about is Quarter, which helps me get insights on the return I’m getting for every click. These tools help you work smarter, not harder. Syllaby and Quarter are great examples of this.

[00:30:27] Nell: Thank you for that, Amanda. I know as a marketer, it can be overwhelming to keep up with the pace of innovation. It often feels like we’re behind if we’re not using every tool available. It sounds like the consensus here is to define where you want to go, use tools to help with research, and go from there. Does anyone else have specific tools they’d like to shout out?

[00:30:59] Ryan: Well, I usually make my own tools. There are language models that can run on your machine, like Llama, which is up and running in five minutes. You don’t have to send client data over the internet, and it’s very flexible. I like to code my own solutions because then I can format it exactly how I want. But you don’t have to know how to code—AI tools can even help with that.

Today, one of my employees had a problem: he had 50,000 to 100,000 keywords and wanted to remove any unrelated to specific topics. I suggested using a custom machine learning model or cosine similarity, and he ended up using Gemini to solve the problem on his own, with code generated by the tool. You don’t need to know how to code, and that’s the beauty of it.

[00:32:23] Amanda: I totally agree. Just ask the questions, refine, and if you get stuck, there are tools—or people like Ryan—that can help you figure it out.

[00:32:35] Nell: All right, got that. Thank you all. My next question is about content creation. I think we’ve spent a fair amount of time on that, so I want to spend the last 10 minutes of the panel talking about the future. Where do you see things headed in the near term? And if you have guesses for the long term, I’d love to know. Where is this ship headed? Lindsay, do you want to kick us off?

[00:32:58] Lindsay: Sure, I’ll start with the short term, and we can branch out from there. Predicting where AI and SEO are heading is a big space. In the immediate future, we’re seeing changes not just in how content gets created, but also in the layout of search engine results pages. Things are shifting, like how often someone clicks on a result versus getting information directly from the search engine. A good example is when “people also ask” became part of the search results—people freaked out. Now, there’s a new segment with AI overviews in Google that’s potentially leading to lower click-through rates, but it’s hard to measure.

Another challenge evolving with this is our ability to attribute and track success accurately. It’s been difficult to trace a perfect line from creating content to impressions, clicks, and sales. With multitouch attribution, the intersection of AI with a cookie-less world, and tracking issues, things are more dynamic. In the long term, our behavior in seeking information and making purchase decisions will likely change significantly. Websites won’t become obsolete, but how people connect with information and brands will evolve.

When we advise small and medium-sized businesses on SEO strategy, we give three main tips. First, at an organizational level, define what success means beyond just rankings or sessions. You won’t be able to perfectly attribute every action. Second, don’t rush to adopt every new technology just because it’s there. Find tools that save time or improve quality, but avoid getting stuck just playing with tools. Lastly, always put your audience first. Every task should positively impact your end user. If it doesn’t, you can skip it—it’s probably low impact or unnecessary.

Those are my thoughts on the smaller landscape, but I’m curious what others think about AI and SEO moving forward.

[00:36:42] Ryan: You covered a ton of great points. I think we’ve been spoiled in SEO for a long time. We used to measure clicks and conversions at the keyword level, and we don’t get that anymore. This is just the next evolution. Bill Gates said in 2009, “The future of search is verbs.” He didn’t mean people would be typing verbs into search engines, but that they want to do something. Back then, if you searched for something like booking a flight, you’d just get a list of web pages with the word “flight.” Now, people want direct answers, like “How old is Taylor Swift?” They don’t want a website with ads and pop-ups; they just want the answer.

We need to stop measuring SEO by clicks and conversions. Every time a client asks me about ROI, I point out a billboard and ask them to measure its impact. AI is the same thing—we hear about blocking AI from crawling, but that just means your competitor’s products will get recommended instead.

In the future, I don’t think typing into a chat box will be the primary use case for AI. It’ll be more integrated into our devices. We already have personalized AI models running in our phones, and I think AI will just be part of search engines, phones, and everyday life. We won’t actively seek out AI; it’ll just be there. But we’ll still need to influence the answers AI gives, whether it’s a product recommendation or event information. It’s just another evolution we have to adapt to.

[00:40:01] Nate: Ryan, you’re right. There was a similar AI hype in the 70s with computers recognizing handwritten numbers, and people thought AI would take over the world. Now, AI will become as normal as big data. Remember when big data was hyped? Now it’s something everyone expects you to be doing. AI will become part of the norm.

What’s interesting is that Google, and search engines in general, haven’t really operated on keyword searches for a long time. Google determines user intent and builds search results based on that, not just the keywords. Our focus on keywords needs to evolve. We should be determining the needs of users, assigning volume estimates to those needs, and tracking our share of voice for them. It’s wild that many companies still brainstorm content ideas without data to back them up. There’s often just a few content ideas that drive most of the search volume, and that’s where our focus should be.

We need to remember that, at the end of the day, it’s about making users happy. We want to create content that satisfies them so much they’re willing to give us their money. It all comes back to the user.

[00:43:55] Amanda: Everyone has hit on great points. I often hear people ask if Google is being hypocritical, and we have to remember that Google is a business. Their goal is to meet their customers’ needs, and we’re part of that system. For your content to surface, you have to help Google meet its customers’ needs. They’re constantly testing and seeing what engages users. If your content isn’t working, you need to figure out why.

We have to help Google achieve its goal so we can achieve ours. It’s about thinking outside the box and solving the problem in a way that aligns with Google’s objectives. Tools like Syllabi can help identify content gaps, but the key is to keep thinking about how to best meet the needs of users.

[00:45:33] Nell:
Work for Liquid. I know.

[00:45:35] Amanda:
I like functional tools that make my job easier.

[00:45:39] Nell:
I wonder if you can. Yeah, Ryan, I think you have some thoughts about that.

[00:45:42] Ryan:
I just saw Joseph’s question when you brought it up, and I’m going to give him an answer he’s not going to like, but I still want to give it. I don’t think it’s very hypocritical, and the reason goes back to intent. If I want an AI answer, I’ll go to AI, Google, Bing, OpenAI, or my favorite right now, Claude. But if I’m going to your website, I want your experience and expertise, which AI can’t provide. It’s a controversial topic in SEO, but it’s essential. AI can’t give expertise, experience, or a new idea that hasn’t been expressed elsewhere. If you’re the only one saying something, AI won’t suggest it because it’s improbable.

That’s why I visit websites, not for AI content — I can get that directly from AI. What’s unique about your site? In computer science and Google, this concept is called “information gain.” What’s new on your website that isn’t everywhere else or in AI? That’s the focus.

Now, this doesn’t mean you can’t use AI. If you want to use AI to rewrite stock manufacturer product descriptions, great — that’s an excellent use case. You avoid having the same descriptions as everyone else. Or use it for personas, intents, or brand voice — all valid uses. But I don’t want AI content from your website because I don’t want to close your cookie consent notice, deal with browser alerts, or autoplay ads. I can get the content directly from Google without that hassle. So differentiate yourself.

[00:47:44] Amanda:
Jordan Grisha had a question many struggle with: how do you show success to leadership without metrics? It’s all about what drives the business — revenue. Decisions are made based on what will make money. Whether it’s SEO, paid, content, or email, it’s all about making money. You must work with leadership to understand their goals and then find ways to attribute your work to those goals. Without that, you’ll always struggle to show your value.

[00:48:54] Nate:
That’s a great point. Unfortunately, leadership are humans too, so they’re constantly figuring out what metrics matter to revenue, and they’re often wrong. Like Amanda said, it’s a constant process of working with them.

I’m an introvert who loves working with clients or in-house teams, but I can’t just say, “Give me a number. Great, I’ll be back in six months with it through the roof.” Leadership goes through iterations of what they care about, and if you don’t adapt, you’ll fall behind.

With the fast-moving trends in technology, the internet, and everything else, leadership rarely sticks to one thing. I wish they could articulate what matters and stay the course longer, but they don’t. So, we have to be adaptable. It’s frustrating, but it’s part of the job these days.

[00:50:31] Ryan:
One of the most insightful leadership calls I had was with C-levels at a Fortune 100 company. Yahoo came in and offered them a Super Bowl placement — a takeover of yahoo.com on game day. The CMO initially said no because they didn’t get clicks or sales from it last year. But when the rep said Toyota wanted it, the CMO immediately changed his mind and said they’d take it.

Even without measurable ROI, they jumped at it because someone else wanted it. That’s a mindset shift we need in SEO. We should focus on visibility, share of voice, and market share, not just clicks and conversions. If we’re not showing up in AI recommendations, someone else is, and we can’t afford that.

[00:51:40] Lindsay:
Yeah, we need to get closer to the business side of things. If you’re offering SEO services, you need to speak in a business voice. It’s not just about rankings anymore. Conversations should start with, “How’s business? What’s happening in your industry? What are your salespeople being asked?” It’s about understanding business concerns and then connecting them to the metrics we track.

The metrics still matter, but the conversation has to start in the broader business context, not just, “I got 1,000 visits from Google last month.”

[00:52:42] Nell:
True, that. Thanks, everyone. I want to be mindful of time. I loved this conversation and am leaving with a lot of action items. We had some hiccups, but there will be a recording with the blog post, and we’ll try to address unanswered questions.

Before we sign off, can everyone share one takeaway for the audience? Amanda, you first.

[00:53:13] Amanda:
My takeaway would be to figure out how to attribute business value to any strategy you’re working on, whether it’s AI or anything else. Show the contribution you bring to the table and how it helps leadership make informed decisions. Feel free to connect with me on LinkedIn for more tips.

[00:53:57] Nell:
Thanks! Nate?

[00:54:00] Nate:
There’s always so much to do in SEO, and that’s the fun part. But I’ll say this: take a few minutes to think about how you can get off the “content treadmill” and focus on creating the best content in the world. There’s a difference between just churning out content and creating something truly exceptional. Give yourself time to reflect on that.

[00:56:07] Nell:
Great. Lindsay?

[00:56:10] Lindsay:
Put your audience first. Focus on what’s genuinely helpful to them and remember they want a human connection. Ryan mentioned earlier that people don’t want AI-generated content from your site — they want to connect with your brand and expertise. Google will reward that, and your business will benefit.

[00:56:36] Nell:
Awesome. Ryan?

[00:56:38] Ryan:
AI is not a strategy, even though clients often ask, “What’s our AI strategy?” AI is a tool or tactic. Use it for the boring, repetitive tasks, not the creative, fun stuff. Don’t ask AI to write poetry while you do laundry; make it do the laundry while you create art.

[00:57:11] Nell:
I find that very comforting — there’s still a big place for the human brain. Thanks, everyone! Lindsay, Ryan, Amanda, Nate, I loved this conversation. We’ll post the recording and follow-up questions with links to your websites. Thanks again!

[00:57:32] Ryan:
Bye now.

[00:57:33] Amanda:
Take care.