Insights with Trevor Stolber
In this episode, Trevor Stolber, an SEO veteran and co-founder of VibeLogic, shared his perspective on how AI is reshaping the SEO landscape. With his wealth of experience, Trevor provided a refreshing take on the challenges and opportunities ahead. Here’s a quick rundown:
SEO Is Not Dead. In Fact SEO is Far from Dead
Despite repeated claims that “SEO is dying,” Trevor reminds us that search remains critical because people will always need answers to their questions. SEO’s role in delivering relevant content isn’t going away—it’s evolving.
AI in Search: Exciting but Risky
Generative AI tools like ChatGPT are changing the game, but they’re not perfect. Trevor explained how these tools often produce convincing but unreliable content. He emphasized the importance of fact-checking and using AI as a tool—not a solution.
What’s Working in SEO Today
Trevor is exploring techniques like vector alignment, which helps content perform better in AI-driven searches. He also recommends shifting from long-form content to short, highly targeted pieces that meet specific user intents, particularly for mid- and bottom-funnel queries.
Use Search Data Beyond SEO
SEO data is gold, but its value goes beyond rankings. Trevor highlighted how search insights can guide strategies across multiple marketing channels, helping businesses better understand their audiences.
Google’s Challenges and Opportunities
Trevor believes Google’s algorithm has become overly complex—almost to the point of losing control. While transparency and consistency are critical, these challenges also open doors for competing technologies to step up.
What’s Next for Search?
Trevor sees search evolving gradually over the next few years, with personalization and conversational elements becoming more prominent. However, he doesn’t expect a complete overhaul in the short term—fundamentals still matter.
Takeaways
SEO isn’t just surviving—it’s adapting. Whether you’re navigating AI integrations, optimizing for vector alignment, or exploring multi-channel applications, staying ahead means combining tried-and-true strategies with a willingness to experiment.
To learn more about Trevor and his work, visit VibeLogic.com or connect with him on LinkedIn. And don’t forget—SEO may change, but the need for quality, intent-driven content will never go out of style.
Video Transcription of The State of SEO in a Post-AI World
Welcome, Trevor . How are you doing? Thanks. You’re really good. Thanks for having me. Cool. If you don’t mind introducing yourself and including your last name, since I got it wrong and just tell us a little bit of your background and what are you up to these days and all that good stuff.
Yeah you did pretty good. It’s better than most people get my name, Trevor Stolber. Actually. It’s 20 years, I think, isn’t it? . Yeah. It’s been a long time. I met you an seo chat that the awesome form that was online when I was getting Transitioning into seo from being a developer And I was learning things and I went there and that’s I think that’s where we met and I really love that forum i’m really yeah, I do.
I miss it. Yeah, . So yeah i’ve been in seo for I think 27 years now So even before google was a thing I was doing seo. So You know, I’ve seen the change where we used to report on multiple search engines and now it’s pretty much Google, although that is changing.
We’ll probably get into that. So yeah, i’ve been doing seo a very long time fairly recently started a new agency called vibe logic with my business partner stefan bajaya. We say that we’re a an agency with a difference, but everyone says that right? Every agency says they’re an agency with a difference yeah, I like to think that we are because we do things a little differently.
So we use search data Which is the demand side. And we identify opportunities, which is the supply side on the sub and we do it in more of a multi channel way. We do traditional SEO. But the sort of landscape strategy stuff we do is much more interesting, we may be getting to some of that too.
That’s awesome. And I’m glad you brought up the word traditional SEO because to me, SEO is changing. Obviously by no means it’s dead. And whoever says that is mistaken, but it is changing. So if you don’t mind covering the state of SEO today for us, I would, from your perspective. Yeah. Yeah, so SEO has been proclaimed dead many times. There’s always a technology or, business trend or practice that seems like it’s gonna kill SEO and it never dies. So I’m not so worried about it as maybe some are. At the end of the day, people have a certain need that they’re trying to fulfill.
That’s why they do the search, that’s why they use the search engine, which is the vessel to get people to what they want. So really, you’re just taking on that challenge and solving that fundamental need that people have. I don’t see many of the current headwinds in SEO as such a big deal as many do it is maybe a somewhat of a transitional period, I’ve seen lots of the algorithm changes.
I’ve seen fads and trends. I feel like I have a pretty good pulse on what would be a real trend and what’s just a fad. The AI stuff is very interesting. It’s definitely very big topic of the moment. I think what’s probably even more interesting, at least to me, and it’s not talked about so much is the actual nature of the search algorithm itself.
And really I just feel like Google’s kind of got itself in a twist. Knee jerk reacted to certain technologies that were seen as challenging it. It’s a very complex beast in itself anyway. It’s an amazing thing that they do. But I would bet that they’re in a position now where they don’t really know how the beast works properly, like sure, at a basic level, the architecture of an algorithm, but.
All the tweaks, modifiers, inputs other sub vector algorithms and everything that that make search what it has become, which was really quite amazing, I think now are probably starting to become a bit of a headwind.
I agree with you as I, I stopped listening to John Miller a long time ago and I respect him a lot. So don’t by any means, I don’t mean to say anything negative about him, but being a former programmer, I know that it’s, sometimes you just coding things on top of code and solutions and top of solutions and patches and top of patches.
And sometimes you start not really it just becomes a beast, like you said, and now with a lot of the AI solutions that are introducing the search results I wanted to see what do you think about the AI and gen AI integrations in the search results? Is that something, is Google shooting itself in the foot or what’s going on there?
Yeah, it’s, that’s very interesting. That’s one of the more sort of interesting Developments in search. And I think many people are concerned about that on many levels and rightly but I think, content still the king, but the kingdom is under attack from AI.
It is back to that general theme, though. We’re still trying to solve a basic need that people have. So they still have that same need to have their information that they’re looking for found to have their query served. It’s difficult to rank for something if you don’t mention it.
Not impossible, but those methods of SEO of making sure that you’re covering the right topics for things people are searching for, that’s still true because the source of the AI content comes from somewhere. And, really it’s being part of that conversation. One really important thing that I think many people miss though, Google’s looking for what’s called information gain.
And so they’re looking to add to things and an LLMs is as powerful as Gen AI is and LLMs are they’re using already existing information to present something which can be considered unique. But you have to ask yourself, what’s the ratio of unique to rehashed unique? To information gain, right? So is it is rehashed existing stuff.
That’s technically unique. Does that really provide information gain? There’s a certain set of queries that it definitely does do very well for, right? But then there’s also the issue where they have these hallucinations and, it still fascinates me. People are surprised when this technology makes something up.
It’s literally designed to do that. It’s generative AI, right? It’s just very fancy predictive text. It’s making up what’s most likely to come next from the words it knows about. Gross oversimplification, it’s designed to make stuff up and so always amazes me when people get bent out of shape when it actually does do that.
That’s what it’s designed to do. So you’ll be very careful with it and I think one of the bigger issues is it does just make this stuff up and it’s very possible. So I think, when people ask chat GPT, for example, make me a strategy on this, it comes back with something that’s very possible and plausible as a strategy, but there’s not much strategizing that’s actually happened.
It’s not relevant. It’s not applicable, but it sounds good. And it looks right. So it passes through a lot of the standard checks that we’re used to as, just as a human reviewer. So I think that’s that’s one of the big issues with AI in, in search and content generation in itself.
Yeah, you identified a very important topic here on how reliable the data that, this gen ai platform like chat, GPT or Gemini provide how reliable it is. And I think part of the fact that PE people need to understand the foundation of ai. Like how does AI work? And there are some courses in LinkedIn learning that I’ve and I personally have taken because even though was a computer science major and we did
take an intro to AI that was many years ago. So I took a refresher course, especially on LLMs it sounds like it could be true, but there’s a lot of stuff that are made up and they’re designed to be predictive and to make things up.
And we should take that information with a grain of salt. Oh, absolutely. And I did something really interesting with it. So I’ve actually found it to be quite a helpful tool and technology for code. It gets me started on a lot of code and it makes me more efficient in that way.
On a lot of projects that would otherwise take me all day to start can take me like half an hour. So it’s very useful for that. I had a particular like networking challenge that I was trying to overcome and I needed to, it’s a bit techy, a bit geeky, but basically the difference, and you’ll know this I’m sure, you’ll know the difference between little endian and big endian and ARM and Intel processors.
And I needed to basically byte flip the code sequence based on which type of hardware we were working with. And I asked it to make me, But essentially a sniffer function, right? That detected what technology was on the end of it and flipped the byte sequence based on it. And it quickly spat out some code and it looked great.
So I ran it and I tested on a system. And I’m like, okay, that’s what I was expecting. Then I tested it on another system that I was expecting to be the other way, but it confidently said, no, this is a big Indian system. Okay. And I went back and forth a little bit with chat GPT and testing this.
And then I like looked at the function. Doesn’t actually do anything and I replied to it in the chat and I’m like, this doesn’t actually do anything, does it? And it’s Oh no, this is just a hypothetical way that you might structure some of this.
I’m like, how much of that stuff exists in production code now that people don’t know about? Cause it looks right. It gives a response that seems right, but it’s not right. That’s a big problem. Yeah. That’s actually a very good example. And I have a simpler. example .
I was looking for a nursery near my house with a special plant and a very rare palm tree that you can only find in Florida and tropical places that in California. It’s really hard to find. And. It said this XYZ, I forget what the nursery was, like it was a few months ago, has it. Mind you that when I went to the nursery, it wasn’t there anymore, there was a restaurant there, so it didn’t exist.
That place had long, like for years had been closed, and so they just made things up. And I was like, and that’s where I had my awakening that yeah, this is a good tool at the very best. If you’re over relying on this thing too much, you’ve got to fact check it. You’ve got to understand how it works and you’ve got to write guardrails and checks and balances to use it.
Yeah, but going back to SEO, I see a lot of Gen AI results in the search results. How is that? I would think that Google has spent many years refining search results, looking at over 200 signals or more, trying to identify what’s a good result. And now here you have Gen AI stuff popping out on top of the other results.
Yeah, and so there’s There’s maybe two schools of thought on that, right? One is like fight against it and one is if you can’t beat them, join them. It’s a little bit to me like when featured snippets came out and zero click searches and people were like up in arms about, Oh, I’m not going to get the click.
I’m not, They’re stealing my content. And now the AI is that on steroids, right? So exactly. So there’s definitely some things that you can do to make sure you’re part of that conversation. And I’ve been doing a lot of research lately on essentially this like vector alignment thing that I think we’ve talked about a little bit.
So from a gen AI perspective, rather than a traditional algorithm ranking perspective, how do you make a certain set of text? More broadly align to a set of queries, right? And so that’s the vector alignment, right? So when someone searches for this grouping of keywords that you’re interested in How do you make sure your content is more likely to rank for that in an ai world rather than in a traditional search world?
So let’s say you have a set of 300 words just a set of words that are 300 words long. How do you make those 300 words come up for more of those queries than another set of 300 words?
And to an extent, you can do that by looking at the likelihood of other words to be associated with them and doing this vector alignment score. And so what I’m now looking at is. A whole slew of keywords. How closely do they align to this block of text and what do I need to do to this text to make it more aligned to the keywords we’re trying to rank for?
So that’s the, I guess they call this GEO now, Generative Engine Optimization or something like that. So that’s definitely one thing. Another thing. Is to look at the types of queries that AI type tends to serve and it serves informational queries a lot more than other types of queries. And that sort of makes sense.
So what that means from an SEO perspective, as much as I love top of the funnel information based content, I do think we need to look a lot more at mid to bottom of the funnel type content, typically shorter form content.
Generally, throughout the last many years, long form content has generally been the desire and the winner for many reasons. But what I think this means now is I think we need much more specific short form optimized content. We need more of it. And this technique that I’m playing with on how to optimize for that vector approach I think is probably the way we need to go.
So still pretty early on in that. I’m not saying I have the answers to this, but I’m definitely have something that looks pretty interesting that I’m playing with.
One thing I do want to talk about is I think this is huge I think it’s not given anywhere near the sort of press that it deserves. Mark Williams cook in the uk recently relatively recently discovered a vulnerability in Google’s one of their APIs, and he found an endpoint that had an exploit that was available, and it was basically spitting out data, and he managed to uncover this.
He was working on the tool, also asked on the API for that, also asked, by the way, great tool, should check that out. What he found was, Google have, they essentially had a leak and I’m not talking about the Google leaks that’s a whole other topic. I’m talking about this endpoint exploit. And what it showed is there was essentially, the way I take it, is it’s essentially a post ranking QA.
And they found a huge amount of data. Even to the point where yes, there actually is a quality score that every domain has. We knew this SEO conceptualized many of these things that have been largely validated by what Mark found.
So I think it hasn’t been unpacked enough yet. I think it’s huge. And so what I see that as, like I say, is this post ranking QA. And I believe that Google is probably using that process. to do machine learning on it, to feed back into its algorithm or into its ranking manipulations.
And this is what they used to use. Human quality raters for, had the quality raters guidelines. Google had an army of quality raters. They would test subtle algorithm modifications and the perturbations that they had in the search results. And rate them and they would use that feedback and basically direct their algorithm development that way That’s all well and good quite a thing to organize an architect, but that’s been the way for a while now I believe we’ve so it’s actually amazing how much computation they’re doing post search I had always assumed and you should never assume but I always assumed that they were doing this sort of pre search and maybe somehow trying to keep track of what types of queries and process it that way.
But it seems not to be the case, or maybe they are doing that, but they also do pretty expensive analysis post search. Is a parameter checked once they’ve delivered that set of search results? So that’s pretty amazing to me that’s really what they were doing.
I didn’t call Mark out on it. I just said, I don’t really buy that they would do this. And one of the reasons was just some of the, in the leaked documents as well. Like some of the programming practices, the naming conventions of variables, the use of values, like they would mix numbers, float numbers and integers and or they would assign a variable type to something that couldn’t add up to that variable type based on the data that it came from.
They would have a negative score, a negative click value and a positive click value when I have a click value could be positive or negative. It made no sense. So there were lots of bad programming practices in there which made me a bit skeptical, but I, Mark showed me the data and like I reviewed it and it’s very interesting.
So that’s what I think that has gone. I think that element of it has gone a little bit crazy to the point where they now don’t really know what’s the true algorithm.
If people continue to click on result three, instead of result to result three, or an equal would jump ahead, which that makes sense. But I think they’ve gone a lot further with this now. And I think I just believe that they’ve probably gone a little bit too far and that’s why they’ve got themselves twisted Yeah, and they tend to do that … there’s gonna be a learning phase for them too, right? So to figure out what’s a good quality and whatnot,. Do you see Google being challenged by other AI search?
I know ChatGPT, working on GPT search there is already a bunch of other ones out there, there’s specialty ones out there, and then they just also platforms like social media platforms are getting better in search, Do you see Google’s dominance being challenged ? For sure, I think it’s happening.
I think we’ve reached peak Google probably a couple of years ago, realistically, if you look at the numbers, they’re still able to manipulate things to keep their top line revenue good and growing. Stefan and I talk about this a lot. I think it’s a decade before Google is not a major player, even if they get broken up, even if AI technology becomes dominant.
I believe it’s a decade before, really search changes beyond what we see as today. Yes, it’s definitely happening, because look at how long it took for Google to get dominance in the search, I remember reporting on keyword rankings on multiple engines. I think that will come back.
Yeah, I do. It’s going to be very interesting. And you look how long Microsoft Explorer took to die, right? Yeah yeah, I definitely think that’s happening. I think different regions might adopt different technologies. There’s different treatments of these technologies in different parts of the world.
People’s , what’s really funny is I heard I had someone using I was just getting my hair cut a few weeks ago and the gal in the shop asked the phone, I’m thinking about moving to this city. Can you tell me why it’s a good place to live? And it generated an answer that told her the place she was thinking of moving was the best possible place she could move.
She took that to mean it searched all the web, and the web says this is absolutely the best place to be, but it just generated something based, so most people don’t really, the casual user doesn’t really recognize this yet, so I think there’s a fairly extensive adaptation period to technology as well, so definitely things are changing, the tide is changing, but I don’t think it goes out quickly.
Yeah, and I think there’s something needs to be said about educating the searchers about that. Hey, you’re okay using searching Ai chat gpt, whatever tool you use. I do it all the time, but you want to verify your information especially for important stuff, even if we Google just because they’re ranked, it doesn’t mean. Doesn’t mean it’s true. No. Yeah, Google was at pains to point out several years ago, they’re not a truth engine. They’re a search engine, right? They’re a search engine, yes. They are providing information relevant to what you queried, and that doesn’t mean it’s true.
And I’ve seen people get bent out of shape on this as well with the suggested search. And this was a neighbor of mine around COVID time. She would type in a specific number very specific thing that probably couldn’t really have actually happened. And she was like, look, see, it’s making it up.
And I’m like, it’s not making it up. It’s trying to find something as relevant as possible to what you searched. Doesn’t mean the thing it finds is true. It’s just finding some relevant information to your query. Google’s not a truth engine.
AI is not a fact engine. Yeah. Before the election, one of my friend that I know is Oh Google is all lefty. If it shows most of the liberal publications as opposed to conservative one. I’m like, no, it’s just because the liberal publications have been around longer to have more content, to have more backlinks.
That’s where they’re ranking, Fox news or whiteboard or all these other, rights oriented conservative publications are newer publications because conservatives used to do more of a talk radio and, but if you look at the backlinks between cnn.com and let’s say foxnews. com, you know how many more backlinks cnn. com has versus foxnews. com or versus Breitbart. And they have a lot more content. Because they were around since the dot com days like before fox news before Breitbart, right? So so that’s why google looks at them authority.
It’s not because google tends to be Liberal or anti conservative . It’s just because they have a lot more content. They have a lot more backlinks, you know that’s actually an interesting. Slant on it as well. People have long talked about bias in algorithms and stuff like that This is where And this is another thing my business partner, Stefan, and I talk about, there’s a decaying factor, a decaying feedback loop with AI content.
So the more AI content that’s created, that now gets ingested by LLMs, right? So these LLMs that built this content are now ingesting the content that they got built in. You only need a very slight bias and you can go off at an angle and you don’t realize you’re going off at an angle and you can go the wrong way very quickly without realizing it.
So there’s definitely that risk in there as well with these sort of LLMs and them ingesting the content that they produced. And so what should users do? Do they verify their searches? You should always double check where and Google my question too, and you’ve got to get to a source ultimately in this, that’s another really interesting thing.
It seems like people have built up authority Google introduced these EAT metrics to make sure that the sources they were citing were credible. And it seems like AI is throwing that all out, right? It seems that way, but when you go full circle with it, because of the stuff I just talked about, it comes back to, you need that credible source to verify.
Sure. The experts are still the experts. The people that have authority still have authority. You should always test at least two different sources. The trouble is most people don’t Yeah, and you see this so many times so many examples of where people just blindly use ai and and they get it wrong.
I go home I don’t have a lot of time between spending time with my family Being tired of my day of working and you know having my home responsibilities The last thing I want to do is double and check and triple check my services. Yeah, it’s time, right? And I think a lot of it has to also be with the industry.
I don’t believe in government regulations, but I do believe the industry needs to have some sort of self regulation as well. There needs to be some sort of standards that we need to adhere to. I think we need to come out of this closed box formula or algorithm world and open up. We’re so much used to you and I have been practicing SEO for over 20 years.
And a lot of things we do is really hacking, it’s reverse engineering Google. We shouldn’t have to do that. The ideal world is Google opening up its algorithms. This is how we do it. And. I’m just seeing a black box and a bunch of people scratching their heads like we have no idea how this thing works.
Yeah. And in a transparency, my point was transparency is going to be one of the key and obviously regulations, although I don’t believe like a Washington DC type of regulations, but more of a industry type of Best practices, I’ve been like the W3 of the W3C, I’ve been an advocate for standards in SEO for a long time.
And it’s, yeah, the argument has always been, it changes so frequently. How can you do that fair argument? But yeah, I just think there’s some standard things that should, I’m a big fan of standardization. And to be honest with you, like when they say seo changes a lot Means of counting and ranking is your chances seo by its by nature cannot change because people are searching for something And it’s delivering contents.
It’s just it’s the fundamentals. The fundamentals are going to be the same It’s just the mechanism the calculations are going to change even then Often happens very gradually. It doesn’t happen overnight. I mean there are occasional Examples of that major Google algorithm updates, but in most cases, those changes very gradual.
And as we talked earlier SEO is not gonna die because we, humans are beings that have a need. We search for food, we search for mates, we search for a purpose. And then you have to search for information. SEO, as you understand it, is going to die very likely.
But I think like you said, this is going to take a while for that to happen. You mentioned long form versus short form content and I have a blog post ranking right now I mean it was ranking as of last week.
They’re on enterprise SEO audits. Now I know SEMrush was bidding on the keyword and PPC. Ahrefs is ranking. That was a blog post where I literally had a few days off. It was one of the last, long weekends. I wrote a really long form blog post and then I used chat GPT to make it even longer.
And then I came back and made it even, I made even more edits to it. And then obviously we did the whole SEO thing on page optimization, off page, all of that. And it’s ranking. To be honest with you I could have written the same blog post in less than a thousand words. Why is it that long form content continues to outrank short form content? Even to this day, where, short form content could be as powerful, even if not more, than the long form content. And most users tend to digest and consume short form content,
yeah. So I guess I don’t know specifically. It sounds it’s like a really a relatively top level commercial intent term. It obviously must meet the need, right? It must meet the intent, the SERP intent, probably. The example I give to people on SERP intent is, if you want to rank an e commerce product page and you’ve got the best SEO in the world, you’ve got more links, you’ve got everything else, but Google’s serving top 10 how to blog posts. You’re not meeting the intent of that search. And that’s part of this vector alignment approach, right? You need to understand how, and we’re doing some fairly advanced intent modeling as well, and not like the classical, information, commercial navigation, transactional, right?
Much more specific niche specific intent. So once you understand that, when you have that in the search data that we have, you can really granularly identify where the needs are and what type of content needs to serve that. My best guess without seeing it would just be that you just really well aligned to the need of the intent as Google sees it.
Yeah. I did my keyword research and phrase research and I think information wise I’m fairly accurate in that post. So I did the best I could, but I could have probably done it in thousand words, not 3000 words.
And that’s not the intent that Google sees it as, even though they’ll happily charge people probably 20 per click for it. That’s a really good point. You intended to answer the question. Yeah, I really wanted to do that and I wanted to show it as a showcase whenever we’re talking to the client It’s hey, you need to be editorial if you want to rank because we work with a lot of clients that like There too salesly in the blog post when they send it to us We usually sometimes we write posts for them.
Sometimes they submit it and we optimize it when they get too salesy, you gotta adhere to brand guidelines. You gotta be the right tone, the right voice. Even now, one thing that’s like a lot of writers seem to get wrong.
Is really long paragraphs even with the advanced NLP understanding that the modern search algorithms have struggle with comprehension of really long strings of text. Yeah, technology has advanced so much. It’s so good. But if just breaking the text up into the right sections is an important step.
You find that’s a really good tip there and so you’re recommending break your paragraphs into smaller paragraphs and smaller sentences. Is that what you’re? Yeah, that can make a difference for readability as well on the web, because it’s if you go to classical word processing, like on a word document, it doesn’t really look out of sorts, but on a web page, it just does because people scan.
I remember it wasn’t too long ago Cyrus Shepard did a study where he changed all of the H1 tags on his blog, or I think 50 percent of the blog posts, to H2s. And at that, but they were styled and rendered exactly the same. And I think back then, classical SEO theory would have said the 50 percent that you changed to H2 would rank lower, but it wasn’t the case.
Didn’t change at all and it’s because of how google renders it So you could render your heading in paragraph tags as a massive heading And then you could render the rest of your body text as an h1 but styled as a paragraph And google wouldn’t yeah because it’s not how it renders and actually one thing I see missed often as well In google search console go and actually live test and then view the rendered output in google search console I’ve seen that. Some people plot robots. txt rules for certain JavaScript libraries or things like that. And then that can break Google’s crawler. Not the crawler they’re rendering. And, that’s how Search Console and Google Search is analyzing the page. I see that often as that’s a good check to do.
It’s a real simple check, but many missed that. I’m really happy you brought it up because I did a test and I don’t have enough data to say this is how it’s gonna be all the time. But I have told a writer that our meta distributions are supposed to be one sentence, not two sentences, because and a short, you don’t, definitely don’t go over 150 characters.
Just very simple. Whenever we have. A really simple sentence explaining the page in a purely editorial way. I’ve seen it to be a lot less likely for Google to rewrite your meta description and that has been my test. Although I wouldn’t call the results scientific because we haven’t really Documented all scenarios and stuff like that, I tend to see, in order to improve our click through rate and search we simplify title tags and meta descriptions.
You don’t have to fill up the whole thing. 155 characters. It could be a hundred character, 120. Make it simple. Make it editorial. Don’t make it salesy. And it’s very similar to what you just said. It’s you don’t want to just stuff, bunch of stuff together, make it simple for the users because of rendering challenges as well.
Yeah, I think people get really bent out of shape with the exact length of titles and meta descriptions. And again, it matters on the intent, right? So there are certain cases where a top of the funnel fairly high level term, you probably want full control over your title and description.
But for more informational type queries longer, more specific long tail keywords you want to be as specific as possible. So for those type, you can go over the length and your keywords and just sacrifice the fact that, it will probably get rewritten more appropriately for that specific query.
Yeah, there’s definitely some trade offs. It’s not a one size fits all. I hate when you see an output of SEO tools and they say, Oh, your title tags too long. There’s many different reasons for why your title tag or meta description should be a certain length.
You made a good point. It’s some control. It’s some control over the real estate that you have there in the SERP, and it’s some control over the click through rate that you have, which is another thing that Google said for many years they don’t use, but we know now, and we knew a long time ago.
I had data going back to 2016, which showed clip manipulation was a thing. So yeah, again, SEO, this is something that people struggle with a lot. It’s not a one size fits all. This is why SEO say it depends, right? And I always tell my clients about this, it depends all the time.
It so often does depend, but you can’t just say it depends. That’s a really low professional barter set. You have to give, for instance, and examples of why it depends and then put it in context for the client. It’s just not, it’s not acceptable just to be like it depends. That’s not okay. Yeah, and this is also another point where you want SEO to be part of your marketing initiative, not the marketing initiatives.
If you’re putting all your eggs in the SEO basket, you have to understand that SEO wasn’t designed to be a marketing channel. The search is also supposed to be editorial, not commercial. The search optimization can do, but it’s not a marketing channel.
Paid search is marketing, right? So we SEOs tend to work with marketing teams or report to marketing leaders, but it’s a it’s not, it’s more of a science than marketing. When I started my SEO agency, I quickly Added other services, like paid search and paid social, because SEO alone, especially in these days, is just not gonna bear fruit for you if that’s all you focus on.
Yeah. My business partner Stefan, and as a SEO, as someone that owns an SEO agency and someone that’s worked for an SEO tool company conductor he says that SEO is the worst thing to happen to search data. And it is because the search data the supply side of the demand equation has so much value and it has so much more multi channel value than just SEO on its own.
I’m not by any stretch saying SEO is a bad thing, but I’m just saying look at the bigger picture, right? There’s so much application there and it tells you what people want and it tells you what’s there. So that’s one of the things that we’re really focusing on now is How to basically use all that data and position a company in a multichannel kind of way.
And that’s an excellent point too. And another thing is in the transactional keywords and if you look at Google interfaces, especially if you’re on mobile, you’re not even going to see the first listing, SEO listing, until you scroll down quite a bit, right? Yeah, it’s terrible. Yeah, you’re gonna see all these paid listings and paid ads and stuff like that so you definitely have su in the toolbox, but as part of your broader thing I do want to ask you a couple more questions before we conclude this.
Are you using any good SEO tools recently, especially SEO AI tools that you could tell that they’re actually helping you? Obviously, we know, or, and I also know like traditional tools like SEMrush are adding AI features. So I wanted to see like in any recent good SEO tools or SEO AI tools that you have utilized.
I’ve got a few, so I’m a big fan of SEMrush. I think you actually recommended SEMrush to me when I was looking at changing my tool. I think you made that recommendation and I took it and I ran with it and yeah, I really liked SEMrush. Another couple of tools I really like, I think I mentioned already, also asked
com that’s Mark Williams cooks tool. What it does is it visualizes the people also ask questions in a really nice way. It gives you a nice graphic. It gives you a CSV output. It’s really cool tool. So I really like that. Another one I’ve tried is SEO stack by Daniel Foley. That’s an awesome tool.
It’s like Google Search Console on steroids. And what it does is it takes a lot of the Google Search Console data, stores it in a large data source. That lets you break free of some of the limits of Google Search Console. In terms of AI tools, I’m not a big fan of AI tools. It looks like they can do something. But it is, it’s not as good as it looks. And because of the way it’s used perceived and positioned they mask what they’re doing and they’re just not that good. I’ve tested a lot. Like I probably buy far too many SAS tools, but yeah, a lot of them are just really not very good. I’m also working on building a tool right now as well with VibeLogic.
I’m building a lot of our processes and analyses into a tool that just helps make us efficient. A big fan of tools. But yeah, I’d say SEMrush also asked an SEO stack.io seo stack.io, I think is a really good one. Another challenge I have with AI tools is that you don’t know where this data has been saved.
And if you’re working with like clients like in healthcare or sensitive data, with, you have to follow regulations. You just can’t use some of these AI tools.
For sure, data integrity is a huge concern. And really early on in OpenAI’s history they leaked some, or some Samsung engineers have put some code in and it got consumed and it got leaked.
You definitely want to make sure you have it in the right box if you have any of those concerns. Yeah. Fantastic technology. Outside of just search applications, there’s a huge amount of applications and even I’ve even been exploring these AI agent things.
So I have actually have built a multi agent framework and I’ve given them just a couple of SOPs and it reports and they just go away and work. Like it’s not it’s not good enough to just set free. But it’s getting there. I definitely think that’s the direction fascinating just for society in the future of work, because I think that’s going to have a bigger shift in search algorithms.
Yeah, I’m excited to see what agents do. I had a problem with one of the saas tools we use and I was emailing back and forth and since I’m also busy And my ADHD and everything like that. I didn’t even pay, I didn’t even know this whole time I was talking to AI. And because the thing, the answers got repetitive.
And we yet to see how the AI agents are going to impact our lives. Again we’re still, even the AI has been around for a long time. The adoption of it in a commercial world is still fairly new. So we have to yet to see how it would, impact our daily lives, how it would improve it, how many jobs it will take away, how many new jobs it will create.
I’m excited, but I’m also curious how What it will do. And if you have any quick anything to add about where do you think search is going? Is it AI agents? Do you have any guesses? Any predictions?
I don’t want to go too far out, so I’ll be fairly cautious here and just say I don’t think it’s going to actually fundamentally change. I think we’ll see a lot more elements. I think it will become quite a bit more adaptive. I’m personalized and it already is. And I think that this conversational search will probably increase.
So there’s also like context conversation optimization, if you want to call it that. How do you structure your content such that it fits the right conversations and answers the conversations that are being asked? And that’s like keyword research on steroids, right? How do you track that?
How do you it becomes like the Starbucks menu, right? How do you know from that one query where someone’s gonna go? So I but again, I just think there are some fairly solid fundamentals to stick to And you can’t go too far wrong. I don’t see in the next few years, search technology changing significantly, even with this AI encroachment in 10 years, I do think the landscape will be a bit different.
But yeah, the next couple of three or four years, I don’t see major changes. Very cool. In 10 years, I’m going to open a sandwich shop because none of the ones I go to put enough meat in it. So I’m going to show America how to make sandwiches. We are running out of time.
If he can just for our audience, tell us where they can find you, what businesses, what are your one or two businesses? I think you said you run a couple of businesses. Yeah. So survive logic dot com is my new company. We’re fairly active on linked in so you can find me on linked in or so. That’s a main forum.
Yeah, check us out. And hit me up on LinkedIn. If you have any questions, I’d love to geek out on SEO anytime. Yeah, I think we need to have you again in future and talk more specific topics because we both are SEO geeks, but also SEO is one industry that’s going to Keep getting impacted by all the new trends and technology.
So it’s good having you on. It was good to catch up with you again . And thanks for coming to our show. Yeah. Thanks very much for having me. Really appreciate it. Take care and happy holidays. Thanks. You too. Okay. Bye.
