Stop obsessing over AI

Taylor says: “Do not obsess about AI. Don't obsess about it.”

Why shouldn't you obsess about it?

“I'm sure a lot of people have helpful suggestions about it, but frankly, it's all so new that it’s on a trial basis. Anyone saying that they know what to do for this environment, it's not true. They think they know what to do, but nobody knows for sure. It's all based on: What am I doing that's working? What are my peers doing that's working? How do we navigate this?

Google has not provided any guidance – and even if they did, who really cares? As we saw with organic search, even if they did provide specific guidelines, those aren’t always true. My issue is more the fact that people feel so concerned about it. For instance, I have so many clients asking, ‘How am I doing on ChatGPT?’, but a study came out yesterday showing that referral traffic from ChatGPT from July 21st through yesterday (August 26th) was down 52%.

AI is not going anywhere, and we're very much going to have to adapt to it, but I do not think search will change very much. We're going to have to adapt to utilising it as a new technology, similar to how we had to adapt to voice searches using Siri and things like that.

One thing that really does bother me is seeing so many people online talking about GEO, or Generative Engine Optimization. To me, that does not exist. If you want to really get down to the bare bones of it, the crawlers – whether they're crawling the robots.txt file or the newer llms.txt file – are still crawling your source code. The source of information is the same.

I'm constantly hearing that GEO is the new thing because the landscape has totally changed with AI search. Typically, the people who say that then go on to present obvious and foundational tactics that have always applied to SEO. For example, ‘We should use schema.’

One thing we've been really focussing on with our clients is making sure that the authors have authoritative bio pages that show their credentials and show who they are. Are the articles fact-checked? However, that all pertains to EEAT (Experience, Expertise, Authority, and Trust), which has been an acronym in SEO for going on a decade.

I'm super excited about the changes. I'm super eager to see how things modify and where we have new opportunities, but if anyone uses the term ‘GEO’, they have no credibility to me. It's the exact same thing. You're appealing to crawlers. Your source code has to be sharp. Whether it's through schema or your header tags, you're trying to send specific signals and understanding through that source code.

I have a client, and his wife is an academic advisor at a very prominent public university. Starting next week, they're giving the students glasses that will help them with various tasks, using AI. That sounds super vague, and I don't know what that means, but my first question was: ‘Can you tell me the model, the make, and everything about it?’ I want to know where it is getting its AI information, and how we appear there.

To me, the adjustment to AI has nothing to do with search. It's more about where AI is going to be prominent. It could be in a pair of glasses. However, I do not think that the way people use search will change so dramatically that anyone should do anything differently.

All of the ranking factors are essentially the same. I could have a page that's ranking number one on organic that’s down 70% for clicks because of AI overviews, but it's also the top-cited article there.”

Is there anything that you have to do differently to optimize for AI engines, or do you think it's too early to consider doing that?

“I'm not saying, ‘Don't change anything.’ My adaptation is less a response to the changes in search and more a response to the changes in overall technology.

If someone were to say, ‘We're trying to rank in AI,’ I would say, ‘Why?’ I feel that what we're doing, as far as SEO goes, should still make us rank in AI. However, as of last month, Google held 83% of the search volume and, combined, all LLMs had 2.2% of the search volume.

Why am I going to dramatically pivot my strategy when it comes to search? I'm not going to react in a panic.”

Do you not have to skate where the puck's going?

“No. I'm the goalie at one end of the ice with the puck. If I shoot the puck, I don't care where it crosses the centre line. What I care about is how we finish at the end.

I genuinely believe that if you're trying to adapt to AI, and you want to be visible in the next era of how people are finding businesses – from Yellow Pages, to search, to whatever's next – it has nothing to do with search. It has to do with, are you pushing your product so that you are the answer? What is the next technology where people are looking for what they want; looking for that answer?

Eventually, search engines may still fuel and power those sources of information, but you need to understand, as AI gets more incorporated into daily life, how people are using it. How has user behaviour changed when looking for an answer or solution? That, to me, is way more valuable. You won't see an enormous return this year, but nobody will. Nothing sustainable will come from adapting to AI overviews or LLMs. What's going to become sustainable is when you have new technologies incorporating that AI.

For instance, in 2005, you could have a website that ranked number one for absolutely everything. Then, 2008, 2009, and 2010 rolled around, the smartphone was introduced, and the way we do everything changed because we were appealing to entirely new technology. I imagine every single person listening or reading is seeing, at minimum, 65% to 35% mobile to desktop traffic. The entire procedure changed.

People are going to keep going to Google for the next 10 years, but Google is going to lose market share – unless Gemini or whatever is incorporated into these new technologies. When you start reaching these kids that are 12, 13, or 14 now, in 10 years, how has technology been incorporated into their lives? I don't think they're going to go to search engines the way we do now.

Don't focus on AI overviews and things like that. Focus on producing EEAT because that will translate across all mediums. Don't appeal to AI just because it’s the hot topic right now. Believe in your strategy, rather than being overly-reactionary.

I'm not here trying to rant; I just think that there's a huge overreaction to the impact of AI on the search engine results page. While the SERP itself is going to lose value over the next few years, as AI is integrated into our daily use, it’s still the king of the castle – and will be in the immediate future.

Especially with AI Mode in Google. If people end up using a pair of AI-incorporated glasses that are somehow connected to Google, they're going to pull that information from Gemini. So, we want to rank for Gemini. For me, it's important to not only focus on how we adapt to where we are right now, because you adapt or you die, but you could adapt one day, and something kills you the next day.

I want to know how I can adapt to be prepared for what's coming in 5 or 10 years, and make sure my clients are the same way. I have relatives who are young children, and they look at DVDs the way I looked at an 8-track growing up. It's not easy to comprehend because we think so differently. What we're talking about today is not going to be relevant in 10 years, in my opinion.”

Do you think keyword research has radically changed because of AI?

“I do not. The emphasis on conversational content has really risen (whether it's Quora, Reddit, or whatever it may be), but you have to keep in mind, Reddit and Google have a $60 million deal. There's a reason why those conversational pieces are more relevant than ever; it's because there's money involved.

That being said, I'm all for it. We want to show EEAT, so if my goal is to show that, my brand should be prevalent in the forums that pertain to that.

When we publish articles, I want every author to have a thorough biography page that they link to. If I have clients who are doctors in Florida, I want to have a piece there that shows where you went to med school, what your credentials are, and everything like that. I've also been having experts fact-check these articles, and demonstrating that. There is a bit more legwork that goes into it, but it will pay dividends not just now, but as this all translates.

At the end of the day, they're crawling source code. If we can tell you and signal to you that we have the best content, we are experts at what we're talking about, and we have experience in what we're talking about, that will serve immediate and long-term dividends more than anything.

My suggestion for 2026 is that you do not ignore EEAT. It will pay you dividends in 2026 and so far beyond, regardless of the way technology heads.”

What do you do if a client pushes back and wants to appear in ChatGPT?

“I always want to make the client happy. I'll look into what's ranking for the competition and how we can get them there. I'm also always happy to send back, ‘Here's how much traffic they're getting. Here's how much traffic we're getting.’

With the 52% referral traffic drop from July 21st through August 26th from ChatGPT, even if they've been citing you, you're not getting traffic from there. As an SEO, it's so nice to say, ‘Hey, your traffic's in the green,’ every month. However, that’s not going to pay my bills. What’s going to pay my bills is the fact that your appointments are up, or your leads are up.

Let me put it this way: have you ever hired a company or made any financial decision based on ChatGPT? Again, LLMs have a 2.2% search share. If a client brings LLM visibility up to me, I'm going to say, ‘You rank number one in everything on Google. That's 90% of the market share.’

If you’re a criminal defence lawyer and someone says, ‘I found you on ChatGPT,’ I probably don't want them as a client. Their life is in your hands. Did they not do their research? Did they not read the reviews? Did they not look elsewhere?

AI is great for asking what the interest rate is today. However, if I'm looking for a lawyer to defend my freedom, what is ChatGPT or any LLM going to tell me that reviews, personal experiences, and case studies are not? That's going to weigh much more heavily.

I tailor each strategy to the individual client. What are their users looking for? If you're spending your entire time trying to show up in AI overviews, the click-through rate on that is like 0.3%. What is the point? That's where the shift happens.

You can spend all the money you want, that you were spending to show up as number one or get the featured snippet, but now you're in a zero-click environment. What's the point? You need to emphasise your personal experience, show that you're an expert, show you have experience, and show that people are happy with your services.

If you're a business, you also want to be showing those reviews and star ratings. We all know that the five yellow stars imply something; whether you even read the reviews or not, you assume people are happy with it. It’s EEAT.

Show you are the best at what you do, and that will pay dividends forever. By panicking or reacting to this change in technology, it's going to change again just as much over the next year, so what will you have achieved? Where are we going to be in 2 or 3 years? That should be the question, and that's what you should prepare for, in my opinion.”

Taylor, what's the key takeaway from the tip you shared today?

“Focus on high-quality content and don't obsess over AI rankings, because the way AI is utilised will change dramatically.

ChatGPT was first rolled out roughly two years ago, and everything has changed since then. Everyone who is adjusting dramatically is assuming that the landscape won’t have changed two years from now.

Focus on being the expert in your area and anticipating the changes in technology because the changes in human use and behaviour will dramatically outweigh the changes in AI involvement, from a search perspective.”

Taylor Kurtz is the Owner and Founder of Crush The Rankings. Find out more over at CrushTheRankings.com.