B (2:57)

Yeah, it's been quite a long and winding road, to be honest. I started off in SEO in 2020 when I built a website to help other people learn how to relocate to European countries with SEO. I got that onto the front page of Google and had quite a lot of success with it. And, you know, gradually, of course, creating content was the biggest bottleneck. Right. And I started to look for ways to make this more streamlined and faster and more doable for a small team. And of course, when ChatGPT was launched in late 2022, it was a total game changer. And when Claude hit the scene, it was even better for us because we just love the way that Claude writes. So I guess that was really the entry point into it. And I just fell in love with really the convenience and the ease that these tools bring to my work. And people started coming to me eventually and asking for help. You do SEO, can you help me show up in AI search instead of just in Google? And there was quite a lot of demand for it, even just in the last six months. So, you know, oftentimes you go where the market leads you. And I decided that this was something people really needed.

A (4:02)

Yeah. And didn't you start going into your YouTube channel, RSAmanthanorth, on some of this content as well? And how did that go?

B (4:09)

Yeah, my YouTube channel, well, I've been stalled on it several times over the last few years, but recently decided to really double down and, and start teaching people not. Not just AI SEO concepts, but also how they can automate some of that busy work that. That goes into this, this whole space. You know, like, it's all very well having a ton of leads coming in from your SEO efforts, but if you don't have processes in place to deal with those, then it can really quickly get overwhelming. So that's one of the things I help people with as well, is, you know, building automations to, to capture all of that so small teams don't get overwhelmed.

A (4:46)

Now, everyone who's listening heard me say that you have a PhD in data science. Do you need a PhD in data science to understand some of the things we're about to talk about? And did you think you would even be doing the kind of work you're doing right now with the PhD that.

B (4:59)

You have to answer the first part of the question? Absolutely not. You do not need a PhD or even a master's degree to understand this stuff. I think for me, academia was a bit of a sidetrack in a way, a very interesting one. But several years in, I realized that staying in the academic world was not going to be for me. I've always been more about getting hands on with the tech, getting there, moving quickly, all of that stuff. And it was just very slow moving for me. And so I just really use this as gaining some important background knowledge that can then go through into my work as a consultant. So it was a valuable experience, but probably not something I would recommend to many people.

A (5:39)

Well, and it does bring you incredible credibility, obviously, when you come to talk about these kind of things on shows like this. Speaking of that, I would love you to share what you see from your lens. Some of the biggest misconceptions are about AI search because we're really here to talk about AI search and a lot of people might have a notion of that, might be maybe a wrong notion. So what are some of those misconceptions that you see when it comes to AI search?

B (6:03)

My goodness, where do I even start? Let's start with this one. That traditional SEO is dead. That is not the case. LLM AI search really shares a lot of overlaps with our traditional search in Google. I noticed recently with a client that much of the content that was showing up for them in AI search was actually blog posts that we'd written together two years ago that was still showing up and still sending them leads. So a lot of those fundamentals are still really important. The differences are more in certain aspects of the structure. Being aware of how people interact with AI search, which is very different from how they interact with Google search, and taking more of a holistic approach to building your authority across the web. So you really need to be hitting multiple different platforms like Reddit, Quora, all the social media, for example, rather than just focusing on a blog. And automation can really take a lot of the of the load off when it comes to that. Another misconception I see is people think that especially when it comes to content creation, that a lot of this stuff can just be automated. Now that we have AI, it's really easy just to churn out content, you know, like hundreds of blog posts a month. I don't think this is the case. Obviously it can be done, but I don't think the quality of this content is going to do any business any favors. To me, it's still really vital to have a human in the loop when it comes to creating SEO content with AI. And that goes exactly the same for content For AI search as well, it's broadly the same thing. So yeah, the humans still need to be involved there. And I think another final thing that I've seen already quite a lot of is certain consultants and agencies guaranteeing that they can make you show up in an AI search. Just like in traditional SEO, you cannot guarantee this. You can do a lot of things to make it more likely, but at the end of the day, it's the models and it's Google that decide who's going to show up in them. And some of that stuff, how they make that decision is still a little bit mysterious.

A (8:15)

I'm glad that you said that because it's true that there's a lot of people that feel like search optimization is dying. I'll confess that I'm one of those people. I don't think it's dead, I just think it's fundamentally shifting. And I also love the fact that, hey, you know, these AI models are pretty sophisticated and they're changing constantly and there's no assurances that you're going to for sure show up, but there are some things you can do to increase the likelihood of it. So I want to logically ask this question. When AI SEO as we're calling it, because we don't have a better phrase right now, when AI optimization, whatever we choose to call this, is done well, what is possible? What are the upsides?

B (8:58)

One of the upsides I've noticed a lot with recently with clients is that leads that come to their site from AI sources tend to have much higher conversion rates and they're also a lot more engaged with the content. I think that happens because think about it, when you have a conversation with an AI, you often get quite deep into the conversation. You explore lots of angles, you go down lots of different rabbit holes with it. And I think if along that journey, a brand is then recommended to you, you're much more in the frame of mind to engage with that brand versus just a very surface level keyword search in Google. So I think that is obviously a big benefit. If conversion rates and engagement rates are much higher. That's important. Also the fact that people get a lot of education from the AI now. So chances are, if they've gone through that phase, they're going to know a lot more about your brand or which brand they want to go with, and they're going to be that much further along the journey towards making that conversion. And the second thing I think is really important about doing AI SEO well is that it gives you the opportunity to control your brand narrative and your reputation. And of course that is incredibly important because if there's negative sentiment out there about you, you need to know about it and you need to take action to reel that back in and to take control of it. So AI SEO can of course help you in that area as well.

A (10:33)

You know, one area, and I love this, one of the areas that so many people listening to this podcast have probably experimented with is the deep research side of AI. Pretty much all the platforms have some vers of this, right? Like Google, OpenAI, Claude, Grok, they all have some, some variation of this thing and it's typically resulting in just a few companies or products being recommended and typically outlined in tables and provided pros and cons and a lot of what used to be done by humans. Because you know, the, a lot of people, when they walk into a traditional business, like a brick and mortar business, typically they've already made up their mind what they want because they've done all the Internet research. Right. So this is actually helping now for more complicated purchases, right? Where you want to work with a vendor or you want to work with a agency or consultancy or service provider. And this is where I totally understand why they would come ready because normally they would have to spend hours and hours doing all this research. It's being accelerated by deep research and in some cases they don't even realize they're using deep research because it just goes out and does the research. And that can be a massive benefit, can't it, to a party that gets recommended amongst a small list of other vendors, Right?

B (11:54)

Definitely. And I was just thinking as you were saying that, you know, reflecting on how I personally engage and with AI tools and choose things through them and a recent experience I was choosing, I think it was like two routers, or routers as you call them, for the home and the one that it recommended to me as the best choice, I went and purchased it without hesitation. I didn't even question that this could not be the best choice for me. So I do think that maybe because of the intimacy of the conversations, you know, just how they go, go so deep and it feels like you're talking to a friend. People tend to trust more of what the, the AI tools recommend. And I think that is going to have much more wide ranging implications beyond just brands in the future. I don't want to go down that rabbit hole right now, but I think it's very important to keep in mind how much we trust what these tools up to us.

A (12:47)

Yeah. And we already understand that's how it works on the Internet without AI. And it makes logical sense that you would use AI as a consultant, for lack of better words, to guide you in an important and critical, important purchase decision. And then by the time you've affirmed some of that stuff with your own research, you're kind of ready to go. Thus the claim that an AI search result, for lack of better words, is actually incrementally more valuable than just a Google search result, because there's a lot that goes behind something like that. So now the logical next question is, all right, Samantha, you've sold me on this value proposition of somehow being recommended by AI. How in the world do we begin? Right. Where do we even start with something like this?

B (13:30)

Well, you have to get dirty and get out there and start asking them questions about yourself.

A (13:34)

Oh, okay.

B (13:34)

What I like to do. Yeah. What I like to do with clients and you know, myself sometimes is just start by asking several different models, what do they know about this brand or what do they know about me? And seeing what they come back with. And this can be very illuminating. In fact, I would recommend everyone just put your name into ChatGPT or Claude and ask it what it knows about you as an individual and see if it's in line with what you want people to know about you. But for brands, I would suggest taking it a step further and rather than just doing everything manually, investing in a search monitoring tool. And there's a few that I've tried and think are good. The three I've tried are Peak, P, E, E, C, A I, tracker, T R, A, K, K R, dot, AI and mentions.

A (14:22)

So wait, real quick, before we get into these tools, I want to back up the train a little bit before we actually use the tools.

B (14:29)

We.

A (14:29)

What types of questions should we be asking, for example, ChatGPT about our business? Do you have a set of basic questions, typically that you would recommend somebody start with if they wanted to? Just like say, all right, let's see how much it knows about me. Like, do you have any simple questions you recommend to get started with for.

B (14:47)

An individual or a brand?

A (14:48)

Either one. You can pick a brand if you want to.

B (14:51)

Yeah, if it was for a brand, I'd recommend. Yes. Start with the basics. What do you know about this brand? What are its key products? How does it compare to its main competitors and who are those competitors? I think that's important. And then you can ask it. You could ask it questions about the products or the services the brand offers about different nuances of those. So, yeah, I think Just, just quite basic really. That's what I would ask. And also see what it comes back with and then respond to what it comes back with as well. You don't know, you don't know what it's going to say. Right. One thing I've taken to doing is that if the response doesn't contain the brand that I want it to contain, I ask the AI, why did you not include X brand? And then it gives me a bunch of reasons and that is interesting. You know, I can say that it might say something like, oh, X brand is a good contender, but I chose A, B and C brands instead because of certain reasons. And it will tell you. And this is, this is again quite illuminating.

A (15:52)

I love that. You could also ask who do you believe the target audience is for this? Because maybe it thinks that you are not the target audience. So for example, if you're looking for a specific company that provides a service, maybe you're a small business and ChatGPT knows this and maybe they target big businesses and maybe there's a reason it was left off, you know what I mean? So that's intriguing. And then what about yourself? What about when you're actually asking it about yourself? Especially if you have a common name like I would imagine there's a lot of people with the last name north out there. Any thoughts on that?

B (16:23)

Well, when I tried it on myself, the results were fairly accurate. But that's, I think because I put effort into growing my LinkedIn. I have, I have my website, samanthanorth.com so obviously it's pulling from sources like this to, to construct that, that information base about me that it serves back. So I think in that, that case, if there was someone else called Samantha north who didn't have all those content assets out there on the Internet, probably she wouldn't show up like I do. I think the personal brand and the authority that you already have online does play a big role in it, at least for individuals.

A (17:01)

Perfect. Okay, so now let's get back. You mentioned that there are three AI search monitoring tools. P, E E, C, A I Peak Tracker, trakkr, AI and then mentions. So, so first of all, what in the world explain what these tools do? Because I didn't even know there were tools that did something like this.

B (17:22)

There's a ton of tools. I mean these are just three that I tried and I found quite easy to use. Basically what they do is they save you having to type in manually all these different prompts. That, that is one thing that they do. They do more Than that they, they track like your mention rate. So let's, that is like, let's say you test 50 prompts and you only show up in five of them, then you want to know what, what that means, right? What that rate is and, and to be able to improve that. For example, they also track which models are suggesting you for which prompts for which queries, which is quite important. And they let you dive into the different pages from your website, the different sources that are surfacing, those answers about you or your, or your brand. So lots of cool things like that. I think the caveat here would be, and this is something I've struggled with a bit, is that none of these tools are going to be entirely accurate. I mean, traditional SEO tools are not entirely accurate, right? Like Semrush and Ahrefs, we've always had that issue to deal with. But that is more of an issue now because there's just so many unknowns and the way we interact with these AIs, I mean, we might talk to it for half an hour, back and forth, back and forth, back and forth before it finally recommends a brand based on what we've asked it to do. So I do feel that this is just the tip of the iceberg and that there's so much more underneath that we haven't really tapped into when it comes to measuring how people engage with this stuff. And I think the coming months is going to be interesting to see how all these tools evolve.

A (19:03)

Of these tools, is there one that you prefer to use or do you tend to use all of them and kind of look for like the overlaps? That's the first question. And the second question is, obviously these tools have tapped into the APIs of all the different AI models. Are they using all the variants of the models? You know how you have, oh, three, you have five, you have all these different models and they're all going to yield kind of different results. Do these tend to like try to send the same prompt through all of them? So first, do you use one? And secondly, do you, do they, do they basically use the APIs of all these? Is that generally that's gotta be how they do it, right?

B (19:39)

That's a good question. So the first question, the simple question, I tried them all. Well, actually they are, these tools get really expensive, that they're more costly usually than just old fashioned SEO tools. So I use mentions at the moment, mentions. So I like the interface, it just makes sense to me. But the other two are good as well. I wouldn't want to have three running at the same time at the moment because of budget concerns. But, you know, sure, if people want to do that, there might be some value in it. The other thing you asked, I don't believe they use all of the models. Note that they do not. So just thinking back to onmentions. Yeah. So I saw it analyzing Claude, for example. It was only looking at Claude haiku. Haiku, haiku 3.5, which is actually a model I almost never use in Claude. So that is actually. Yeah, there's food for thought. There's ChatGPT. Last time I used it, it was just using 4.

A (20:35)

Oh, probably 4o.

B (20:36)

Yeah, 4o, the really common ones. And that's another interesting thing that comes to. You've just given me another direction here. The majority of people, you know, they're probably searching on the free versions, right?

A (20:47)

Yeah. So we should probably set up some free versions to run tests, right?

B (20:51)

Yeah, but like power users, I guess, like us, we're searching in all kinds of different paid models. So how do monitoring tools account for that difference? I don't know. I don't know the answer to that.

A (21:03)

Well, and it's fascinating because there is this side of it which is like, hey, ChatGPT and Claude have memory now, which means they kind of know who we are, they know what we're interested in. They're becoming almost like Google search. They're highly customized to us. So in an ideal world, we're going to want to use like in an Incognito window in Chrome or something, the free models. Right. Just to see what an unbiased version of the AI would likely search up to someone who's on a free model. And that's really intriguing and just adds the complexity of what we're talking about here, isn't it?

B (21:36)

Yeah. And I'm thinking now that that's, that's, that's probably another reason to use a search monitoring tool because, yeah, I think that would get past the memory issue. Right. Because I don't see how that could be tapping into the memory of you that your chatgpt had. I don't think it could.

A (21:52)

No, it couldn't.

B (21:52)

So that isn't another reason to use them rather than manually. But if I was going to do it manually, yeah, I guess I would log out of everything, get on incognito, use a VPN for whichever region I was interested in, like try to make it forget me as much as possible. But I haven't gone down that road because it seems like too much effort when there are tools like this, that can do some of the heavy lifting at least.

A (22:12)

Yeah. I'm just thinking how I would do it, even though I've never done this. And obviously that's the free plan, the free way to go. And here's what we know, folks. ChatGPT is like 95% of the population uses ChatGPT. And then after that you probably have Google, Gemini and Claude probably tied for second place. And do you also recommend Google search also now that it's becoming a little bit more AI centric with its AI, whatever they call it at the top, The AI overviews? Yes.

B (22:39)

Yeah. I mean, somehow, you know, it sounds weird, but I tend to ignore them because I've noticed quite a lot of inaccuracies, really glaring inaccuracies for some things in those.

A (22:49)

What about other people that might be using it though? What about the other consumers who may be looking to it as their answer now?

B (22:55)

Proceed with caution.

A (22:57)

Yeah, you want to see if you're being listed though, right? You want to maybe ask the same thing and do a Google search query.

B (23:02)

Yeah, I mean, to see who comes up in the top 10 or in AI overview.

A (23:06)

Like if you're asking for. I don't know, I'm just ideating with you. But if you're asking who are the top three in this niche and you're hopeful you're one of them, you might want to see if you come up in the AI overview, my guess is. And compare that up against maybe Gemini. Yeah. All right. We've done. We've gone down a techie.

B (23:23)

I have to throw in one more Mike.

A (23:24)

Yes.

B (23:24)

Perplexity.

A (23:25)

Oh, yeah.

B (23:26)

In my research, ChatGPT is the biggest referrer or brands. But often perplexity is in position too. So I'm talking like Google Analytics, like who is sending traffic to a website, which models and chatgpt followed by perplexity. Actually, Gemini doesn't show up that much. And I think that might be because at least for Google Analytics, Gemini and some other models do not send referrer data. So you can't always identify that they are the ones sending you traffic, which.

A (23:53)

Is bizarre because it's Google. You would think that they would send that the UTMs or something, you know.

B (23:58)

So it's a moving thing. Shifting all the time.

A (24:03)

Is your marketing falling behind? In my 15 plus years of tracking marketing trends, I've learned to spot the signals early. And right now the pace of change in marketing is rapidly accelerating. What used to take years is now happening in months. AI capabilities that seemed impossible last year are already Outdated. And I'm concerned that marketers who aren't prepared for what's coming next are going to be in a bit of a bind. The tactics that you're using today, the strategies that you spent years mastering, even the way you think about your audience, it's all evolving a lot faster than everyone realizes. This is exactly why social media marketing world exists. When you attend our 2026 event, you'll be positioned to lead AI strategy conversations in your company and for your clients. You'll be able to master tools and tactics before they become mainstream. And you'll also build authentic connections with future ready marketers that will have a big downstream impact on your business. Don't wait until everyone else figures this out. Join the marketers who want to get ahead and stay ahead. We have a really big sale going on right now. Visit social mediamarketingworld.info and grab your tickets today. Let's navigate this change together. Okay, so we start with understanding the landscape. We've been spending a lot of time talking about asking the different AI models questions about your brand or your personal brand. And we mentioned all these different monitoring tools. Once we've done that, the logical next thing to explore is how can we increase the likelihood that we do show up at the top or amongst the recommended results inside of AI?

B (25:47)

Yeah, so there's a couple of phases I like to go through on, on a typical project with this and it probably makes sense to attack it in this order. So what I would do first is if a, if a, if a brand or a client already had content on their site that was showing up in Google Search, then I would take an audit of that and look for the top performing ones and see if any of those are also performing in AI, SEO, AI search, and then kind of attack them from a AI search optimization standpoint. So I would do stuff like change the headings to put them into the form of questions that mimic how people might talk to their AI.

A (26:31)

Yeah, real quick, before we get into the questions, how do we know if something is coming up in search, whether it's coming up in AI search? Are we specifically asking AI to give citations so that we can cross check this? Because you know how normally when AI results come up, they don't necessarily have any unless it's perplexity or deep research. They don't normally have links to the sources. So do you have a tip on how to ascertain that the AI is using that particular source as its training material?

B (27:00)

Yeah, there's a couple of ways to do it so you can use one of the tools like mentions and they will show you which URLs are feeding the prompts. Or you can also just, if you have Google Analytics set up, you can just look at what sources on ChatGPT, like narrow it down to ChatGPT and perplexity and look at what URLs are sending, are receiving traffic from those sources from your site.

A (27:23)

Okay, so at this point we are auditing our content that we have created on our website because that's where we have access to our analytical data and we're looking specifically for traffic coming in from like Perplexity and, or chatgpt or any of the other models, frankly. And then when we find that now we're at the point where you were about to talk about changing the headlines and stuff. I just, I wanted to back that train up a little just to help everybody understand like, like the connection point there. So please proceed with how ought we maybe to modify this content that's already ranking in AI?

B (27:57)

Yeah, so I like to change these basic or sometimes meaningless headings into more question based formats to mimic more how people will be interacting with, with an AI tool. I think that's this is important. I usually put a summary box at the top of each article now because that gives models something very quick to grab from. So it makes their task easier, I suppose, which is important. I would make sure that the article is using pretty simple language. So no jargon, no complicated sentence patterns or structures or anything like that. Get rid of all that stuff and learn about schema markup. It's, that's the code that goes into the header of your page and it tells bots what your page is all about. So I'm not an expert in schema markup. I still use ChatGPT to help me construct them. But for example, if you had a homepage, you could use an organization schema to make sure make it very clear that your homepage is about an organization. If you had a How to post, you could use how to schema to tell it that. So this really makes it crystal clear. These were always kind of important, but I feel that in traditional SEO you could get away with skipping them a lot of the time. But for AI SEO it's becoming more and more important to give it that clarity on what the page is about.

A (29:18)

All right, so let's say we have a About us page that's really detailed or there's an article that's on our website that talks about our services. First thing you recommend is taking some of those H2 or H3 headers, which are those like paragraph headers, you know, that are bold, typically above a big section. And instead of having it be like Social Media examiner services, change it to what services does Social Media examiner offer? Or something like that. Right. And the reason why is, do you want to explain like what the rationale behind that is?

B (29:50)

Yeah, it's more likely that a person would type into ChatGPT, what services does Social Media examiner offer? So it's just mimicking that directly.

A (29:59)

And then the box at the top that you referenced, this summary box, what type of information do you recommend going into a box like that? And can AI help us craft that information?

B (30:10)

Definitely. I always use AI to help me. It's like a tldr, you know, too long, didn't read kind of thing. So yeah, just a summary of what you think are the most important points of the article, what you would want this article to get picked up for. But you can also easily use Claude or ChatGPT. Just put your whole article in there and ask it to pull out the five most important points for a summary box. It does that very easily and pretty neatly.

A (30:39)

When we are altering a piece of content on our website, how long do you think it takes for the AI models to re scrape, for lack of better words, our content? Here's my understanding of it. Models are up to date as to a certain date, typically. Right. Like they'll normally say this model has been trained on information ending on July 30, 2024. Hypothetically. Right. So is it true that we have to wait until the model updates before that stuff works its way back into the brain, if you will, of AI or what's your thoughts on this?

B (31:13)

Yeah, that is, that is a good question. And I believe that if, if the model is accessing the Internet, then that training date cutoff point doesn't really matter anymore. I remember in the past, like Claude was always the tricky one because they took ages to add real time Internet search to that and it was always constrained by its last training date. But yeah, if it's accessing the Internet directly, then it shouldn't be an issue in terms of how long does it take for your new content to hit the models? I don't really have a clear answer to that. Again, it's probably the old chestnut, it depends like it used to be with SEO. But my understanding is that it, it is often faster than content hitting the front page of Google, so there is that.

A (31:56)

Okay, my next question, which is kind of in this vein, but it's not something that, that I've asked you in advance is the model that these tools are using to do the search. OpenAI has a big investment from Microsoft and I would imagine Microsoft Bing might be the search model behind the scenes that's powering the actual search when it goes out there on the web. And obviously in Google's case, it might be Google Search. Right. And in Claude's case, I have no idea. Do you think it's safe to assume they're using traditional search to find the answers when they're going out on the Internet, or do you think they've developed their own search? And where I'm going with this is like, if, should we go to Bing and see if the results show up in Bing and then cross check to see if it shows up in ChatGPT? Do you understand where I'm going with this?

B (32:47)

A little bit? Yeah, yeah, yeah, yeah. I have an interesting little story to share on that, actually. It relates to Bing, so. So I always focused on Google Search Console and I kind of ignored Bing for the longest time, right until I got wind of the fact that I should be taking advantage of Bing because. Because chatgpt is what you just said. Exactly what you just said. It's. It's likely pulling from Bing. So what I did was I got my site and I went and made sure it was fully registered in Bing Search Console with the site map and all of that stuff in there, just like with Google and click submit and waited. And then I think within days in Ahrefs, I started to see that my site was getting more citations from ChatGPT. So it's very coincidental, but the timing was definitely significant there quite soon after. So I think, yeah, don't ignore Bing. We do tend to overlook it a lot because Google has had such a monopoly on this stuff. But I would recommend if you have a website, get its sitemap and get it all set up in Bing Search Console as well as Google and you might be pleasantly surprised with the results. So, yeah, like factor Bing into your strategy.

A (34:01)

Those of us that are Google focused, which is most of us, what do you have to do to set up Bing Search Console? Like, that's new to me. Do you just Google Bing Search Console and you're done in a minute.

B (34:10)

Google Bing Search Console and create a login. Follow the instructions to submit your sitemap. It was pretty straightforward. Like, I don't remember the exact steps, but if you, if you Google it or Bing it, we don't have to.

A (34:23)

Add any special Code on our website or anything like that, presumably. Right.

B (34:26)

I don't remember adding special code. No, I know for Google Search Console it was a little trickier because there were special codes to add. I remember the Bing process was quite straightforward, so I couldn't say 100%. It was a while ago now, but yeah, it was not difficult.

A (34:43)

Yeah. Because it does logically make sense that these AI tools are going to leverage existing search scripts for lack of better words that are out there, so they don't have to reinvent the wheel, you know, in Perplexity. I don't have any sense as to whether Perplexity is using Bing or Google or whatever, but I would imagine maybe in their case they have developed their own query model or whatever. But I thought that was an interesting question. I'm glad I went down that rabbit hole with you. Okay, so so far to this point we've talked about step one is to query the models and to see how much they know about you, how much they know about the business that you work for. And then step number two is to actually audit the content that is showing up somehow inside of AI. And one tip you talked about was using Google Analytics to look for traffic coming from ChatGPT and perplexity and maybe Claude or whatever models and then to go to that content and to potentially modify that content and see whether or not that actually makes a difference in showing up more inside of the AI models. And what else? Because obviously some of us don't have a lot of content, right. That's on there on our website. Maybe it's other entities that are writing about us or referring to us a little bit. Or can we make new content and somehow influence the likelihood that that that new content shows up inside of AI?

B (36:10)

Yeah, absolutely. Well, you mentioned that other entities might be referring to us. That can also be a good thing because that's showing that we have authority outside of just our little blog, our little corner of the Internet. So yeah, like that aside, let's talk about creating new content. That's what we can control. Highly, highly recommend doing that. So I typically work with software companies, SaaS companies, and for them I recommend creating content like what is the best X tool for Y task or X competitor versus our brand? Things like that, like direct comparisons. These are important. Maybe also alternatives to X competitor. So like discussing a list of possible alternatives because these are all queries that people would make when they're quite far down that purchasing pathway, Right?

A (37:00)

Oh wait, does that mean we should say X competitors to us? Like we may as well own that. Right. Like, let's say you're software company A. You could write competitors to software company A and you could be the one that ranks and then you could selectively choose the competitors if you wanted to. Is that kind of what I'm hearing.

B (37:19)

You say we could do that? I mean, I like to keep this very, very factual and objective because you, you don't want to be. Because there's always this innate bias. Like if somebody comes to your site and you have best email marketing tools for small business and yours is clearly number one. That's obvious. You're going to put it number one. It's not going to be number 10. Right. But you also have to be really objective and factual and clear about why you think you're better than the competitors. I've seen lots of brands do this. Well, some do it badly, but it can be done gracefully and elegantly and, and the right way.

A (37:54)

Well, yeah, and I'm thinking one way you could do it elegant. No, this is very important, I think.

B (38:00)

Yeah.

A (38:00)

One way you can do this very eloquently is if, you know, you're, for example, for small business and the competitors are for big business. Well, then you could clearly call out like this. Our software, you know, is for small business and this is for big business. And if you put a table together and you're a big business, then you're obviously not going to choose us. Right. You're going to choose the bigger one. And you have to acknowledge the strengths of the competition here in order for this to be, from a human lens, valuable. But what about from an AI lens? Are they going to read it no matter what?

B (38:33)

Yes. Well, another thing that's important here is that the AI models are trained to be balanced. Right. So, yeah, what you just described is gold. That is what you should be doing as well. You should be presenting things in a balanced way. You can't claim that your brand is the best at everything because that is unbalanced and doesn't make, doesn't make sense. No one's believing that stuff. But if you do that and say that, you know, we are good for these use cases for, for this, this icp, but here's a competitor that's also good in this area, that's stronger in this area, it's better fit for this particular use case, that kind of thing is, is Goldust, because yeah, that's what they're going to be picking up in their, in their search for, to serve us balanced content for our queries when.

A (39:15)

We write these Articles. Are we publishing them on our own website? I think we are. Right. So they're going to know it's from us, presumably. Right.

B (39:23)

Yeah.

A (39:24)

We're not like launching a brand new website called Best SaaS Tools for Blank or are we? I don't know. What's your thoughts on this?

B (39:30)

Well, that would be more of an affiliate play, I think.

A (39:32)

Right, I see. Okay.

B (39:34)

Yeah. So if you're a brand then, yeah, like definitely have lists on your site but you, you have to be like, like don't try and be something you're not. Don't, don't pretend that it's not. You're not going to put yourself number one. You've got to own that. But do it in an elegant way.

A (39:49)

When you are writing these articles, do you typically recommend they be very lengthy and detailed or more precise and short? Because these are potentially going to rank not just in AI but also in Google Search. Right.

B (40:00)

They should say everything they need to say without fluff. So you don't want to be padding it with unnecessary things. Right. Precise and to the point is better than fluffy and long winded just for the sake of it.

A (40:11)

Got it. Okay. And then when you're comparing yourself directly against a competitor, I can imagine there's multiple ways to do this. One way to do this would be the way that most marketers would want to do this, which is a very good head to head comparison where you call out features that you know they do not have. Generally when people do this, they do not call out the features that the competitors have that they do not have. And it's because they're using it as a, as a closing tool. So do you recommend two variations of this, one for the AI and one for the customer, or do you recommend just being fully transparent because you might face a little bit of an internal battle? You know what I mean? Inside the business.

B (40:50)

That's a question. I mean that is a heck of a question.

A (40:54)

I mean from your purchase, someone who's advising people, what do you tell them?

B (40:57)

I would say for the AI models and for the public facing content, you should be transparent because people are not going to be fooled if they come to your site and they see that, that they expect you to put yourself up there and big yourself up. But you also have to do it in the right way for your internal sales team, like do what you think is best, do what it takes to sell. But of course I always advocate transparency as personal as my personal philosophy on it. But internal teams, I expect they have their own philosophies as well.

A (41:29)

Love it. Okay, so beyond creating new content on our own website, are there any other ways we can influence the AI results on sites perhaps we do not own.

B (41:39)

So many, so many options for this. So the most influential ones at the moment are Reddit and Quora. So both of these are gold mines for so called unbiased perspectives on things. You might have heard a while back that Google noticed that people were typing in like best software for X and then putting Reddit appending Reddit to the end of that query. Right. And they were doing that and I've done it myself because we're looking to bypass all the affiliate stuff that's on page one of Google and really get an honest, unbiased opinion on a site like Reddit. And Reddit is very stringent on its no promotional content stuff. So you cannot overtly post stuff that looks like affiliate or promotional content. You've got to tread very, very, very carefully. So that factor, that, that feature of Reddit makes it very powerful in terms of having unbiased recommendations for the models and I suspect they've picked up on that and that's why they're using Reddit as such a powerful influence. So yeah, definitely try and get your brand into conversations on Reddit, but you've got to tread very, very carefully. Like don't just go barging in there talking about your brand because you're probably going to get banned and deleted. So that one is something that tech research.

A (42:58)

Yeah, so it sounds like we don't have a lot of control over that one. So let's move to some that we do have a little bit more control.

B (43:03)

Over our LinkedIn personal profiles, for example. You know, LinkedIn is, is amazing for, for brand visibility and authority. So what I've been doing for the last three years or so is posting almost every day about different problems in my niche, observations, thought, thought leadership posts, all kinds of things, sometimes just a blatant. This is my offer. Here's what it contains once in a while, but not that often. It's mostly about getting into the conversation, building authority and engaging with, with people in my space and using that as a way to, to attract inbound leads, which it does quite well, I have to say. It's, it's a very, very powerful tool for that. And what all of this does in relation to AI search is it builds like this foundation of authority, brand authority across the web. So it's no longer just about your one little blog. It's about are you a known quantity across the Internet, across Multiple different social media platforms across Reddit and Quora, across forums. And also in terms of like media articles as well, like another factor that's really important is trying to get mentioned in, in the press if you can. So signing up to. It's not HARO anymore, is it? I think it's called sos. Source of sources. Peter Shankman. I get his email.

A (44:27)

Yeah, I know who Peter is, but I, I think I might have known that he launched something new. I had forgotten it was Help a reporter out.

B (44:34)

Yeah, I think it's called SOS now. I'm definitely getting his emails a few times a day and there's a lot of juicy stuff in there. And if you can pitch to journalists and get them to cover you and mention your brand or your name or whatever in the article, then that again is fodder for large language models and it reinforces your authority in turn.

A (44:53)

You know what's interesting about LinkedIn and a lot of people may not know this, but there is a lockdown happening on the social platforms. Meta is has locked down its platforms because it has its own model called Llama. So you can no longer like ChatGPT and Google can't scrape the stuff that happens on Meta. And then of course you've got. LinkedIn is one of the few platforms that's actually opened, but it is owned by Microsoft and they could decide to close it down. But it's one of the few platforms that allows third party AI models to come in and scrape it. And that's why a lot of people find a lot of value in LinkedIn because it's the way ChatGPT right now learns a lot about who you are as a business person. And X of course is locked down because it's owned by Elon. You know what I mean?

B (45:43)

I don't use X anymore.

A (45:44)

Yeah, but, but, but there's all these social platforms. YouTube is probably locked down, I'm guessing just by Google. Right. So you're beginning to understand like the, these sources of information are getting locked down right now. As of today, LinkedIn is one of the few that's not locked down and all the AI models, models are using it for information. But eventually if it does get locked down, well, then you're going to have to find alternative places. Like you had mentioned Medium also. Do you have any experience with Medium at all?

B (46:12)

Yeah, I do, but, but I also want to add something to what you were saying there. I believe that there's still value in building a presence on these other platforms for another reason. Not not just directly, because the, because AI tools might be scraping them, but also because you are just growing visibility and like awareness of your brand. Right? So let's say if somebody has seen your brand on Facebook and Twitter and YouTube, even if it's not being scraped directly, they still have knowledge of it. So if it gets recommended by an AI tool, then it's going to click in someone's head and they're going to be like, oh, I've seen that all over the place. I feel like I trust this. I feel like I know it already. So I still believe there's value in being active on socials, even if they're not directly being scraped anymore. Even if they are this walled garden.

A (47:03)

I'm with you on that completely. And some of these people are going to end up writing about these things on their blogs or they're going to be talking about them in their podcast. And that data is also obviously, yeah, talk to us about Medium real quick.

B (47:14)

Medium. I mean, yeah, I haven't got too much to say about it other than to me it's quite a useful and low effort way to, to spread my content further. So what I do now is I just really like copy paste my blog posts onto, onto Medium. I don't even change them a whole lot. I just put them out there. And this is more of like throwing things at the wall and seeing what sticks exercise, because I don't really know if it's brought me any benefit yet. But Medium is a powerful platform and there's a lot of domain authority in medium.com so I'm hoping in a kind of slightly parasitic way that that might be useful for me. And it takes minutes to do, so why not? You can even automate it, right? Publish a new blog post on WordPress automatically publish it to Medium as well. It's not hard to do.

A (48:04)

Awesome, Samantha. This has been a fascinating exploration of AI search as we're calling it. If people want to connect with you on the socials, do you have a preferred platform? And if they want to check out your business, where do you want to send them?

B (48:18)

Yeah, I absolutely prefer LinkedIn. So it's Samantha north on LinkedIn. Feel free to connect with me. I think it's open for that. And I also have samanthanorth.com which is where I blog about marketing, related AI automation and search topics. My business is @emigre AI, which is emigre AI. And you can also put forward slash SME and get a nice freebie to help you do better in AI search, like a checklist for the things that you should be fixing on your site to make it show up in AI search. So please go ahead to emigrate AI SME and download that and hopefully it helps you out.

A (49:01)

Samantha, thank you so much for sharing your insights with us today.

B (49:04)

Thanks for having me.

A (49:04)

Mike hey, if you missed anything, we took all the notes for you@socialmediaexaminer.com and be sure to follow the show on whatever app you are listening on. And if you've been a listener for a little while, we would love a review and we'd love it if you would share with your friends. You can tag me on Facebook, LinkedIn and or X which are the platforms I'm most active on. And do check out our other shows, the Social Media Marketing Podcast and the Social Media Marketing Talk Show. This brings us to the end of the AI Explored Podcast. I'm your host. Host Michael Stelzman will be back with you next week. I hope you make the best out of your day and may AI help you become more successful.

B (49:44)

The AI Explored Podcast is a production of Social Media Examiner.

A (49:51)

This is the year to finally come to Social Media Marketing World 2026. Grab your tickets right now by visiting social mediamarketingworld.info.