Clients are starting to ask: “How do we show up in ChatGPT?” If you don’t have an answer, this on-demand webinar is for you.
In the webinar, Partner Growth Manager, Cara Corbett, covers:
- An introduction to AI Search (AI SEO, GEO)
- How AI Search is changing the path to purchase
- Why you shouldn’t ignore Google
- Making sure AI is talking about your clients the right way
Scroll down to watch the recording.
Webinar Transcript
00:00:11.181 –> 00:00:16.181
Hey, everyone! Can everyone see me? Put it in the chat if you can see me.
00:00:16.181 –> 00:00:25.181
Okay, amazing. So I just… I thought that this was recording the entire time. What’s a… what’s a webinar without some technical difficulties, am I right? Okay.
00:00:25.181 –> 00:00:36.181
So we’re gonna restart. Hey, everyone. Um, very sorry for the delay and the technical difficulties. The good news is, is that I’m actually only going to be talking for about 30 minutes.
00:00:36.181 –> 00:00:45.181
So we’ll still have ample time for some questions at the end, but let me get set up with my screen share here.
00:00:45.181 –> 00:00:56.181
And we’ll get started. Okay… Can everyone see my screen?
00:00:56.181 –> 00:01:14.181
Please… yep. Okay, awesome. All right, let’s get going. So, hey everyone, sorry again for, uh, for that, and thanks for your patience. I’m Kara Corbett, the Partner Growth Manager at CISO Digital, and welcome to our webinar, AI is Talking About Your Clients. Make sure it’s Saying the right thing.
00:01:14.181 –> 00:01:33.181
During this presentation, we’ll share a little bit about the progression of AI search over the last year, what we’re actually seeing work to get brands recognized, valued, and mentioned in LLMs like ChatGPT, Perplexity, and Gemini, and also introduce our new AI SEO service so that your clients can start proactively pushing their AI engine presence.
00:01:33.181 –> 00:01:53.181
There will be a small question period at the end of the presentation, like I just mentioned as well, where myself and our managing director, Andrew, will answer all of your questions. So feel free to drop those in the chat at any time, and we’ll get them at the end. And Andrew, if there’s any other technical difficulties, please, uh, please jump in and let me know, because I am not monitoring the chat.
00:01:53.181 –> 00:02:04.181
All right, um, we’ve entered a world where AI isn’t just answering questions, it’s our decision-making partner across the purchase journey, and shaping brand perception and visibility.
00:02:04.181 –> 00:02:11.181
Llms like ChatGPT, Google’s Gemini, and Perplexity are the ultimate influencers now, and they’re talking about your clients.
00:02:11.181 –> 00:02:17.181
Whether your clients are part of the narrative, or being misrepresented entirely depends on how they show up in these models.
00:02:17.181 –> 00:02:22.181
Whether your clients are part of the narrative, or being… I just said that, sorry.
00:02:22.181 –> 00:02:28.181
We’re not just witnessing a shift in search behavior, we’re watching the entire concept of search being redefined by models.
00:02:28.181 –> 00:02:46.181
These LLMs, they don’t behave like Google. They don’t index the web and spit back a ranked list of blue links for users to click through. Instead, they interpret and they influence. They’re trained on vast behavioral data from what you buy on Amazon to what you post on Instagram, to how you phrase a query in ChatGPT.
00:02:46.181 –> 00:03:02.181
And these aren’t just new algorithms, they’re new decision engines. So AI SEO is a different game to traditional SEO. It’s not a bolt-on, it’s a total rethink of how your client’s visibility, brand perception, and conversion paths work. In the age of model-driven discovery.
00:03:02.181 –> 00:03:08.181
So these are four questions that every brand is probably asking you right now, and that they should be asking.
00:03:08.181 –> 00:03:12.181
And this is exactly what our new AI SEO service is built to answer.
00:03:12.181 –> 00:03:34.181
So question one is, what is AI saying about me and my competitors? While the model already has an opinion, the question is whether it reflects your client’s reality. We uncover how models like ChatGPT currently perceive your client’s brand, and how that stacks up against their competitors. Are they seen as an authority, a recommendation, or just missing entirely? Now, that’s the starting point.
00:03:34.181 –> 00:03:42.181
Number two, can AI actually access my existing content? It’s not just about what’s written on the webpage anymore, it’s about what the model can understand.
00:03:42.181 –> 00:03:48.181
We analyze your content to see if it’s structured in a way that LLMs can ingest, summarize, and use.
00:03:48.181 –> 00:03:53.181
Because if the model can’t parse it, it won’t recommend it, no matter how good it is.
00:03:53.181 –> 00:04:01.181
Number three, can I seed content and answers? We don’t game the system, we train it. And this is where the true opportunity lies.
00:04:01.181 –> 00:04:06.181
By publishing the right kind of content in the right places, we can help your clients seed facts.
00:04:06.181 –> 00:04:17.181
Phrases and associations that influence how the model talks about them and their offerings We’re correcting misclassifications, we’re filling gaps, and we’re building brand perception from the inside out.
00:04:17.181 –> 00:04:33.181
And number four, what are the business outcomes? We track how often your brand is cited in AI overviews, we measure shifts in how models describe you versus your competitors, and we monitor content adoption, keyword association, and entity relevance inside of LLM outputs.
00:04:33.181 –> 00:04:45.181
And we connect that to the outcomes that matter. Increases in branded search, higher conversion from AI-exposed users, reduced reliance on paid ads, and growth in share of model inside AI-powered platforms.
00:04:45.181 –> 00:04:52.181
So, in short, we don’t just help you optimize for search engines, we help you earn trust in the evolving frontier of search.
00:04:52.181 –> 00:05:01.181
So by the end of this webinar, our goal is for you to be able to answer these four questions and provide real solutions for your clients.
00:05:01.181 –> 00:05:16.181
But first, let’s look at some of the data. Ai SEO is already reshaping the way people discover brands, and it’s happening right now. Search demand for terms like AI SEO has exploded, and Google Trends data shows that it’s at its peak.
00:05:16.181 –> 00:05:23.181
And why? Because clients, your clients, are now asking, how do we show up in Google’s AI mode or in ChatGPT?
00:05:23.181 –> 00:05:33.181
And they’ve realized that traditional SEO visibility isn’t enough anymore. It doesn’t guarantee inclusion in these AI-generated answers, and that’s what’s driving the urgency.
00:05:33.181 –> 00:05:43.181
Now, we’re optimizing for the worldview of large language models here, for the way that they reason, rank, and recommend, and that’s a completely different challenge and a massive opportunity for agencies.
00:05:43.181 –> 00:05:50.181
And the agencies that understand this shift early are going to be the ones defining the market, not chasing it.
00:05:50.181 –> 00:06:03.181
And if you’re looking at raw numbers, Google is still really dominating in traditional search compared to other popular engines, but when we actually compare it to ChatGPT in 2024, it did 373 times more global daily searches.
00:06:03.181 –> 00:06:19.181
No surprise there, but it’s important not to be fooled by stats like this. Firstly, ChatGPT is growing super fast, and these numbers are probably already massively different. Secondly, Google searches include huge quantities of basic navigational searches.
00:06:19.181 –> 00:06:22.181
Asking about the weather, for example, and even just quick facts.
00:06:22.181 –> 00:06:28.181
And these AI models are increasingly getting more valuable decision-influencing and making searches.
00:06:28.181 –> 00:06:45.181
Openai has half a billion weekly searches. Um, and they’re not browsing, they’re deciding. These aren’t traditional searches, they’re opinions and recommendations filtered through a model’s understanding of your brand and market. And when that model misinterprets your positioning, it affects real buying decisions.
00:06:45.181 –> 00:06:53.181
So don’t get distracted by legacy metrics. This isn’t about replacing search, it’s about recognizing that the influence layer has moved upstream.
00:06:53.181 –> 00:06:59.181
Models like GPT are on the rise, and this graph is gonna look very different in a couple of years’ time.
00:06:59.181 –> 00:07:09.181
And ChatGPT’s growth is explosive. It’s gone from $300 million to 800 million monthly users since the start of this year alone, and Perplexity is seeing similar growth.
00:07:09.181 –> 00:07:15.181
Having more than tripled their active monthly visits between March 2024 and March 2025.
00:07:15.181 –> 00:07:24.181
Now, that’s a paradigm shift in user behavior, and people aren’t just googling anymore, they’re asking, they’re exploring, and they’re actually making decisions inside of these models.
00:07:24.181 –> 00:07:36.181
Chatgpt is now a mainstream decision engine used by Gen Z, by students, by CEOs and marketers, and increasingly, your clients’ customers as well. So every time someone asks, what’s the best product or service for me.
00:07:36.181 –> 00:07:46.181
Chatgpt is making that call based on what it’s read about brands. Every page, every product, every review, it all feeds the model’s understanding.
00:07:46.181 –> 00:07:53.181
And the model is already recommending, it’s already influencing, and it’s already deciding. And that’s why brands need to act now.
00:07:53.181 –> 00:08:06.181
And user search behavior is becoming more complex. Chatgpt makes deeper research much easier than Google, as it gives an instant answer, but people are still using both traditional search and AI search, but for different reasons.
00:08:06.181 –> 00:08:12.181
So according to SEMrush, AI search is primarily used for complex informational search queries.
00:08:12.181 –> 00:08:17.181
Noting that 61% of searches showed that this, uh, informational intent.
00:08:17.181 –> 00:08:24.181
Whereas Google is mostly used for simple navigational queries, where people are looking to get somewhere, looking for something specific.
00:08:24.181 –> 00:08:32.181
So what does this tell us? Users are switching to LLMs to find information, but using Google for basic navigation.
00:08:32.181 –> 00:08:43.181
And things are transforming every single day. Ai search is now changing the path to purchase, even. With OpenAI already incorporating products and shopping features into their search platform.
00:08:43.181 –> 00:08:51.181
On April 29th, they announced ChatGPT Shopping, and it’s not a vague experiment, it’s already rolled out across free, plus, and Pro users.
00:08:51.181 –> 00:08:58.181
So what does it do? It curates products based on user prompts, it shows visuals and prices and reviews.
00:08:58.181 –> 00:09:01.181
And it links directly to where you can buy the product.
00:09:01.181 –> 00:09:06.181
And these aren’t ads, although they aren’t very far off, they don’t look very far off.
00:09:06.181 –> 00:09:24.181
They’re model-driven recommendations, which means your product shows up if the model thinks that you’re the best choice. And that’s a completely different game than bidding on keywords. You really have to earn that relevance in the model’s mind So the brands that understand how to shape the model’s perception today are gonna be the ones that show up tomorrow when someone says.
00:09:24.181 –> 00:09:39.181
Find the best product for me. So it’s easy to get swept up in the hype around ChatGPT and perplexity and meta-AI, but let’s be crystal clear, make sure you don’t ignore Google, because this is still the dominant force in search.
00:09:39.181 –> 00:09:54.181
And with Gemini, Google has been quietly but aggressively rebuilding search behind the scenes from the inside out. It’s no longer just links, it’s AI overviews, it’s AI mode, and it’s Gemini being embedded across search, Gmail, YouTube, Docs, and ads.
00:09:54.181 –> 00:10:06.181
So think of like this, um, Facebook’s market share has dropped, but it still rains. Even with TikTok on the rise, Facebook continues to dominate and influence in infrastructure, and Google’s really in that same position right now.
00:10:06.181 –> 00:10:11.181
So what was Google’s response to ChatGPT and all the other LLMs?
00:10:11.181 –> 00:10:21.181
It was AI Overviews in AI mode. So, AI Overviews has been around for about a year, and it’s known for giving AI-generated summaries that appear on the top of Google search results.
00:10:21.181 –> 00:10:28.181
Before the traditional listings, providing users with a quick overview on a query, and pulling from multiple sources online.
00:10:28.181 –> 00:10:38.181
So understanding the format of these is really key to reverse engineering visibility. Most overviews are made up of short, scannable paragraphs, often paired with lists.
00:10:38.181 –> 00:10:47.181
According to Surfer SEO, 78% of all AI overviews include a list of some sort, and this can be ordered or unordered, or both.
00:10:47.181 –> 00:11:00.181
And why? Because it’s about information retrieval at speed. Lists are fast and very scannable, and they align with how models structure responses and also how users skim and parse information for value.
00:11:00.181 –> 00:11:09.181
So you’ll especially see ordered lists when the prompt includes any kind of step-by-step or how-to, which now makes up a massive chunk of AI-powered search results.
00:11:09.181 –> 00:11:15.181
So, if your content isn’t structured clearly, if it’s dense, if it buries answers.
00:11:15.181 –> 00:11:34.181
You’re actually training the model to skip you. Ai mode has been rolled out in the US as of last week And the rest of the world probably isn’t far behind, um, but this expands on what AI overviews can do with the ability to explore further, with follow-up questions and helpful web links, very similar to perplexity in ChatGPT.
00:11:34.181 –> 00:11:45.181
And we’re seeing that influence is shifting from the search engine results pages to the LLMs, and that winning an SEO doesn’t necessarily mean that you’re guaranteed to win an AI SEO.
00:11:45.181 –> 00:11:59.181
They’re playing completely different rules, and the data is proving this. So, for example, recent research from VARN shows that 79% of the URLs featured in AI overviews don’t actually appear in the traditional top organic results for the same keyword.
00:11:59.181 –> 00:12:10.181
So, in other words, you could be ranking number one on Google and still be completely absent from the model’s response. And this is a fundamental divergence in how visibility is being calculated.
00:12:10.181 –> 00:12:27.181
So what does that tell us? Ai overviews are operating on different logic, they’re not just scraping the top 10, they’re building answers based on relevance to the model’s understanding. It’s a different algorithm, a different system, and a whole different game, one that requires new tactics and frameworks, and a new thinking.
00:12:27.181 –> 00:12:32.181
So let’s look at a bit of a difference between how traditional SEO and AI SEO compare.
00:12:32.181 –> 00:12:48.181
Traditional search engines are really the gatekeepers of visibility, and this involves ranking in search results, getting clicks to a website, optimizing the dwell time, and improving the user experience. It’s really manual, and it’s built around how people are browsing.
00:12:48.181 –> 00:12:52.181
They can… they control access to visibility, and they don’t recommend, they just rank.
00:12:52.181 –> 00:13:04.181
But AI SEO is a different game. It’s about building trust with the machine, not the person, and the goal isn’t necessarily to get someone to click It’s to get the model to recommend your client’s brand or products.
00:13:04.181 –> 00:13:11.181
And that means we’re optimizing for clarity, structure, and interpretability, because these models, they’re not reading like humans.
00:13:11.181 –> 00:13:18.181
They’re scanning, they’re summarizing, and they’re interpreting. So if your content is messy or unclear, the model’s going to skip you.
00:13:18.181 –> 00:13:26.181
Traditional SEO relies on manual research and intent mapping, where users are actively clicking on links to find what they’re looking for.
00:13:26.181 –> 00:13:34.181
Whereas AI SEO leverages automated summarization and model-driven ranking logic by doing the clicking and the research for you.
00:13:34.181 –> 00:13:52.181
In SEO, the aim is to provide the best possible experience for the end user, so if the UX on your website is poor, people are gonna leave. But with AI SEO, summarization and data clarity is what’s more important. And the aim is to feed the LLMs with the content they need to recommend to users.
00:13:52.181 –> 00:13:59.181
Seo offers broad traffic through the blue links, and AI SEO focuses on narrow, high-intent moments.
00:13:59.181 –> 00:14:04.181
Also, with AI SEO, you need to shape your message so the model understands it first.
00:14:04.181 –> 00:14:14.181
Because that’s what gets you surfaced in the first place. So in a nutshell, they both share DNA, but the approach and the strategy and the outcome are very different.
00:14:14.181 –> 00:14:23.181
Your clients are aware the game is changing. They’re often not just looking for SEO anymore, and they have questions you need to answer, or they’re gonna buy AI SEO from somebody else.
00:14:23.181 –> 00:14:31.181
So this is what clients are really thinking. What does ChatGPT say about me? Can the model even see my content? And can I influence what it recommends?
00:14:31.181 –> 00:14:36.181
They’re asking how AI sees them, and how they can claim their share of model.
00:14:36.181 –> 00:14:51.181
So when we talk about share of model, we’re talking about how LLMs perceive and prioritize your clients’ brands relative to their competitors. It’s not just about search rankings anymore, it’s about how often and how confidently these models surface your clients in responses.
00:14:51.181 –> 00:15:02.181
Across relevant industries. This is about brand presence and AI-generated answers, how often the model chooses your client over another when someone asks a question in your industry or category.
00:15:02.181 –> 00:15:10.181
So the key question becomes, is the model aware of your brand? Does it trust it? And is it choosing it to feature it when it matters most?
00:15:10.181 –> 00:15:18.181
And that’s the essence of the share of model. It’s a new kind of visibility metric that captures your client’s share of the conversation over the time.
00:15:18.181 –> 00:15:35.181
So how can clients claim their share of model and make sure that AI is saying the right thing about their brand? And that’s where we come in with our new service, AI SEO, because bridging that gap between what clients want and what the model needs is exactly what we’re offering. We don’t just optimize for rankings.
00:15:35.181 –> 00:15:47.181
We optimize for how models think, interpret, and decide. We help to make sure your clients show up when someone types the best product or the service for me. In Gemini or ChatGPT.
00:15:47.181 –> 00:15:54.181
And we built an AI service to capture and answer each of these four questions, making it as easy as possible for you to sell to your clients.
00:15:54.181 –> 00:15:58.181
So, number one, what is AI saying about me and my competitors?
00:15:58.181 –> 00:16:11.181
We start with market monitoring, analyzing how large language models describe your clients, how often they’re mentioned, and how they stack up against competitors. This really gives us the baseline for brand perception inside the model.
00:16:11.181 –> 00:16:26.181
Number two, can it access my existing content? That’s where AI-ready content optimization comes in. We audit the content for clarity, for structure, for accessibility, and model readability, making sure it’s designed to be picked up, summarized, and recommended.
00:16:26.181 –> 00:16:43.181
Number 3, can I seed content and answers? We achieved that through new, regularly published AI-optimized content that deliberately triggers the signal LLMs are trained to recognize. Clarity, originality, authority, and association with relevant topics.
00:16:43.181 –> 00:16:55.181
And number four, what are the business outcomes? And this is where it all comes together. We track analytics, traffic, and attribution. We monitor AI referrals, branded search lift, and content impact.
00:16:55.181 –> 00:17:03.181
So you can see how visibility and models translates into real-world performance. And this is how we turn AI SEO from theory into measurable value.
00:17:03.181 –> 00:17:11.181
So, I’ll now go through some of the tactics that we employ as part of our service that answer these questions.
00:17:11.181 –> 00:17:17.181
So, for market monitoring, the first step is to understand what AI is actually saying about your clients.
00:17:17.181 –> 00:17:24.181
So how are they being mentioned in comparison to their competitors? How often are they being cited as sources for relevant user queries?
00:17:24.181 –> 00:17:29.181
So what you’re looking at is a screenshot from our own proprietary tool that we’ve developed in-house.
00:17:29.181 –> 00:17:35.181
And it essentially looks at how all of the LLMs are mentioning your client’s brand and their competitors.
00:17:35.181 –> 00:17:41.181
So how does it work? Well, we do a lot of the legwork and the heavy lifting by feeding the system with hundreds of relevant prompts and phrases.
00:17:41.181 –> 00:17:51.181
That clients will want associated with their business. We then monitor how many times each of the LLMs mention their brand, and that’s how we measure the share of model.
00:17:51.181 –> 00:18:00.181
Every asset is now a brief to someone else’s model, so if you want these models to understand you, you need to be legible to them.
00:18:00.181 –> 00:18:23.181
That starts with the basics, structure and accessibility. Because LLMs, they’re not patiently waiting for JavaScript-heavy pages to load. They actually favorites that are clean and easy to render. And accessibility, it’s not just about compliance, it’s actually about clarity. So semantic HTML, alt text, metadata, all of it helps models understand the meaning of your content.
00:18:23.181 –> 00:18:34.181
Not just the words. So yes, accessible design is very inclusive, but in the age of AI, it’s also a strategic advantage, because if the model can’t read you, it can’t recommend you.
00:18:34.181 –> 00:18:46.181
So SUSO can help ensure your clients’ websites are technically sound, so that their content is immediately available to AI crawlers, and not hidden behind layers of front-end complexity.
00:18:46.181 –> 00:18:56.181
Um, LLMs like ChatGPT and Gemini scrape and summarize content in real time. They don’t queue, and they don’t wait. So, if the model can’t read your pages fast.
00:18:56.181 –> 00:19:13.181
You don’t exist, they’re gonna move on. If it skips your content, it’s gonna pull the answers from a faster competitor instead. So, site speed, it’s not just about UX anymore, it’s also about model accessibility. And that’s the difference between being cited in an AI answer, or being completely ignored.
00:19:13.181 –> 00:19:19.181
In AI SEO, speed equals visibility. And if you’re not fast, you’re not found.
00:19:19.181 –> 00:19:32.181
So our technical team can help audit your client’s site performance and suggest recommendations such as compressing large images or using a caching plugin, on how to improve loading times for your site.
00:19:32.181 –> 00:19:44.181
And another thing we can do is LLMs.txt, and we like to think of this as the robots.txt of the AI era. Um, but instead of guiding search engine crawlers, it’s really built for LLMs, telling them what they can and can’t access.
00:19:44.181 –> 00:20:00.181
How to interpret it, and how to use it. So this is your way of shaping how the model sees content proactively. And here’s why that matters. Clear structure means cleaner citations, better summaries, and higher chances of being surfaced in AI-generated answers.
00:20:00.181 –> 00:20:08.181
Is it widely adopted yet? No, but that’s the point. Implementing it now puts you ahead of the curve, while everyone else is still playing catch-up.
00:20:08.181 –> 00:20:27.181
So this isn’t just about technical hygiene, it’s strategic visibility engineering for AI-first discovery. The models are already reading, the LLMs.txt just helps you make sure that they’re reading it the right way. And we can take care of creating the LLMs.txt, making sure that the AI models can find the right content and recommend your clients.
00:20:27.181 –> 00:20:35.181
And if you want AI to understand your content, you definitely need to label it clearly, and that’s exactly what structured data or schema markup does.
00:20:35.181 –> 00:20:46.181
It’s essentially a piece of code on the back end of the website that gives models like Gemini and ChatGPT the context that they need to interpret your content accurately. You’re not just throwing information on a page.
00:20:46.181 –> 00:20:57.181
You’re speaking the model’s native language, and this results in better indexing, richer SERP features, and higher chances of being cited or summarized by AI in real-time responses.
00:20:57.181 –> 00:21:04.181
The aim here is to train the model to understand who you are, what you do, and why you matter by marking up your content with relevant schema.
00:21:04.181 –> 00:21:15.181
So, for example, with blog posts, we can mark them up with article schema, which allows you to provide LLMs with that contextual information about the author, so that they know exactly who wrote the article.
00:21:15.181 –> 00:21:22.181
This is something our team audits, as well as identifies opportunities for implementations with this service.
00:21:22.181 –> 00:21:33.181
We also create content that models recognize as more relevant for their next recommendation. Ai platforms don’t just pull from the most authoritative source, they also pull from the clearest one.
00:21:33.181 –> 00:21:41.181
In this example, we can see exactly which sentence Google extracts to build its AI overviews to answer the question about when the best time is to have a protein shake.
00:21:41.181 –> 00:21:56.181
It’s not always the most details, it’s the most digestible. And that’s why your information architecture matters more than ever. The way your content is structured, like sentence length, layout, and clarity, directly affects whether the model can summarize and reuse it.
00:21:56.181 –> 00:22:12.181
So here’s what we do to help. Use short, direct sentences, answer the core question immediately, avoid keyword stuffing because the model sees right through it, focusing on long-tail, high-intent queries, keep content fresh and relevant, especially around trending topics.
00:22:12.181 –> 00:22:33.181
And this is about creating content that the model recognizes as more relevant to its next recommendation, because every single piece of content you publish is either training the model to trust you, or to ignore you. Remember, we’re not just writing for rankings anymore, we’re shaping the model’s perception, and that perception determines who shows up in that AI-generated answer.
00:22:33.181 –> 00:22:45.181
Ai wants to understand your articles quickly, so that it can determine their relevance to the user’s query. So, if your content doesn’t front-load the value, like in the example on the screen.
00:22:45.181 –> 00:22:49.181
The model is less likely to pick it up, and then users are also likely to bounce, too.
00:22:49.181 –> 00:23:03.181
And that’s why key takeaways, or TLDRs. Um, those are really useful, and those sections, uh, come in very useful. They’re not fluff, they’re not signal. Um, they’ll tell the model, here’s exactly what this page is about.
00:23:03.181 –> 00:23:13.181
Here’s why it matters, and here’s what to pull. So whether it’s a short paragraph or a bullet list, giving the model a clean entry point helps it index and summarize your content more effectively.
00:23:13.181 –> 00:23:27.181
And that’s the kind of structure that makes your content more recognizable and reusable and recommendable. So like we’ve said before, we’re not just publishing content, we’re creating the right content that’s going to make a change in how the model sees us.
00:23:27.181 –> 00:23:35.181
Traditional SEO is about keyword targeting, but with AI-powered platforms, they care more about how people ask, not just what they ask.
00:23:35.181 –> 00:23:44.181
And that’s why conversational content wins the AI game. Your content needs to feel natural, intuitive, and aligned with how people actually speak.
00:23:44.181 –> 00:23:54.181
We’re not just writing for crawlers anymore, we’re writing for models trained on human dialogue, and this also includes Reddit threads and support tickets and even comment sections. So what works?
00:23:54.181 –> 00:24:02.181
The aim is to use full, direct answers to real questions that offer complete thoughts and uses language that mirrors the user.
00:24:02.181 –> 00:24:12.181
And we can see in this example that they’ve done exactly that. And ChatGPT has taken a negative information about cardiovascular endurance from this conversation sentence and use it within its response.
00:24:12.181 –> 00:24:22.181
And this is what makes your content recognizable to the model, not just crawlable, but usable. Because AI search, it’s not just about being indexed, it’s about being the answer.
00:24:22.181 –> 00:24:32.181
Ai isn’t just summarizing the internet, it’s also curating what’s useful and distinct, and that means repeating what everyone else says is not going to get your clients seen.
00:24:32.181 –> 00:24:40.181
Llms are actively scanning for original signals, which includes unique perspectives, standout features, proprietary insights, and first-party data.
00:24:40.181 –> 00:24:45.181
And those are the nuggets that make the model stop and say, this is worth including.
00:24:45.181 –> 00:24:54.181
So we need to clearly articulate what sets your clients apart, what do they offer that no one else does, and what unique story, stat, or capability positions them as the go-to brand.
00:24:54.181 –> 00:25:05.181
And when we feed that into structured, scannable content, and back it with the schema where relevant, we’re not just improving visibility, we’re improving the model’s understanding of your authority.
00:25:05.181 –> 00:25:10.181
And that’s how you go from being another option to being the recommendation.
00:25:10.181 –> 00:25:33.181
Llms and search engines are getting better at asking, who said this and why should we trust them? And they’re looking for authorship, credibility, and a track record. And that’s why your clients need This to be in order to claim their share, their model share. Um, they have the expertise, but they need to make it visible and verifiable, and so that means real author bios with links to LinkedIn or relevant profiles.
00:25:33.181 –> 00:25:43.181
An author schema markup, so that models can connect the dots. It’s really not about telling your life story, but signaling the authority in the way that models can read and trust.
00:25:43.181 –> 00:25:55.181
So take fitness advice, for example. Models want to recommend the version written by real fitness or medical professionals, someone who’s published before, has credentials, and shows up consistently as an expert within the space.
00:25:55.181 –> 00:26:03.181
And this is how you position your content to be recommended, not just indexed. So, you don’t just want to show up, you want to show up as an authority.
00:26:03.181 –> 00:26:17.181
And think of AI as your most diligent researcher scanning everything, looking for proof of value. It’s not just reading content, but it’s also trying to decide whether the content it finds can be trusted, and whether it should recommend it.
00:26:17.181 –> 00:26:23.181
So, showcasing your wins through awards and case studies and client testimonials are great authority signals.
00:26:23.181 –> 00:26:31.181
They show the model that you’ve delivered real results, and when you highlight these clearly on your site, you make it easier for the model to understand.
00:26:31.181 –> 00:26:37.181
Um, so understand your niche, uh, recognize your credibility, and also cite you with confidence.
00:26:37.181 –> 00:26:43.181
Um, this is about training the model to see you as a proven solution, and not just another option.
00:26:43.181 –> 00:26:48.181
So make your success stories easy to find, clear to parse, and impossible to ignore.
00:26:48.181 –> 00:26:59.181
Conversations about your clients are happening everywhere, in Reddit threads, Quora Answers, blog comments, LinkedIn posts, news articles, review sites, and even industry forums.
00:26:59.181 –> 00:27:05.181
And these aren’t isolated mentions, they’re data points. And LLMs are picking up on all of it.
00:27:05.181 –> 00:27:12.181
They’re trained to read across the web and understand what a brand is known for, who’s talking about it, and whether it’s trusted in its space.
00:27:12.181 –> 00:27:20.181
Even unlinked brand mentions help form a perception. So if your client’s brand name keeps showing up in relevant, high-quality conversations.
00:27:20.181 –> 00:27:35.181
They’re signaling authority without needing a single backlink. And that’s how the model learns what your brand is associated with, what you’re credible on, and whether to include you in its responses. And that’s why off-page activity matters as well. So things like guest appearances.
00:27:35.181 –> 00:27:43.181
Thought leadership, forum contributions, and brand mentions in niche publications. These all feed into the model’s worldview of your brand.
00:27:43.181 –> 00:27:48.181
So for this, our strategy is simple. We want to present your clients where the conversations are actually happening.
00:27:48.181 –> 00:27:54.181
Because the model is listening, and every mention helps shape what it thinks of you.
00:27:54.181 –> 00:28:00.181
And when happy customers leave positive reviews, they’re also not just influencing other prospects.
00:28:00.181 –> 00:28:16.181
They’re influencing the algorithms as well. Llms and search engines actively analyze sentiment across platforms like Google, Trustpilot, Yelp, and niche directories, and they’re using those reviews to decide, is this business trusted? Are their customers happy? And therefore, should I recommend them?
00:28:16.181 –> 00:28:30.181
And in this context, positive reviews become real-world proof. That your product or service delivers value. And the effect? More trust from the model, better positioning in AI-generated recommendations, and more visibility at those high-intent moments.
00:28:30.181 –> 00:28:41.181
So what’s the play? Proactively ask your satisfied customers to leave reviews, prioritize platforms that matter for your industry, for example, if you work in the travel niche, TripAdvisor would be advised.
00:28:41.181 –> 00:28:48.181
And then showcasing that feedback on your website, where the model can see it, parse it, and factor it into its perception.
00:28:48.181 –> 00:28:55.181
An AI SEO, a happy customer is more than a conversion. They’re also a credibility multiplier.
00:28:55.181 –> 00:29:05.181
And you can amplify your AI signal with digital PR. Digital PR is not just about backlinks anymore, it’s about visibility in the places that LLM models trust.
00:29:05.181 –> 00:29:16.181
Um, so when a brand is featured on authoritative sites mentioned by credible sources or quoted in industry press, all of those signals tell AI that this brand matters, and it’s a recognized voice in the space.
00:29:16.181 –> 00:29:25.181
Llms pick up on these mentions with or without links, and use them to gauge relevance, trustworthiness, and authority. So strong digital PR does two things.
00:29:25.181 –> 00:29:38.181
One, it brings in real human attention, but it also amplifies your AI discoverability behind the scenes. And this is how you scale your share of model, by appearing in the conversations and publications that shape how LLMs perceive you.
00:29:38.181 –> 00:29:47.181
So SUSO can help execute digital PR campaigns that get trusted sources to talk about your clients so that AI listens.
00:29:47.181 –> 00:29:59.181
And this is where AI SEO becomes measurable and not just theoretical, in that analytics, traffic, and attribution section. So we don’t just help your clients show up in AI-generated answers, we also help them track it over time.
00:29:59.181 –> 00:30:09.181
So we set up AI traffic attribution inside Google Analytics so they can see exactly how people are reaching their site through platforms like ChatGPT, Perplexity, and Gemini.
00:30:09.181 –> 00:30:15.181
This gives us real visibility in how the content is performing inside LLM ecosystems, and not just in traditional search.
00:30:15.181 –> 00:30:22.181
And with that data, we can refine the AI SEO strategy, we can double down on what’s working, and we can fix what’s being ignored.
00:30:22.181 –> 00:30:37.181
And then tie it all back to qualified traffic, engagement, and business impact. Because at the end of the day, it’s not just about being recommended, it’s also about being discovered by the right audience, and then measuring the value of every AI-driven touchpoint.
00:30:37.181 –> 00:30:43.181
Now, let’s see how we employed the same tactics to grow our 3D printing client’s AI overviews visibility.
00:30:43.181 –> 00:30:53.181
With some of these tactics, we’re now tracking over 90 key business phrases for the brand within Google’s AI overviews. And then here’s the result in action. For PETG temperature resistant.
00:30:53.181 –> 00:31:05.181
This client appears in the AI overviews and holds the featured snippet. It’s a double whammy dominating both the traditional SERP and the AI search result. And that’s what AI SEO is designed to do.
00:31:05.181 –> 00:31:14.181
So this is where it all comes together. We know that the landscape is shifting, we know that AI is now shaping how people discover brands and make decisions and form trust.
00:31:14.181 –> 00:31:26.181
And your clients are asking the right questions. How do we show up in AI-generated answers? How do we stay competitive is the way people search evolves. And by partnering with us, you can help your clients win in this new AI-first environment.
00:31:26.181 –> 00:31:31.181
So, this is a new complete, uh, completely new service that we’ve built for the AI era.
00:31:31.181 –> 00:31:37.181
Every brand’s needs and every model’s perception of them is different, so we don’t have off-the-shelf packages yet.
00:31:37.181 –> 00:31:50.181
We offer bespoke pricing tailored to your client’s current position and their goals, but to give you a sense of entry, $3,000 is going to get you the fundamentals, or the foundation needed to start shaping visibility in AI platforms.
00:31:50.181 –> 00:32:07.181
From there, the more they invest, the more we can unlock. Deeper optimization, ongoing content seeding, entity refinement, and AI performance tracking. So you can offer this as your, uh, as a one-off project to your clients, or as an ongoing monthly retainer, depending on how aggressively they want to move.
00:32:07.181 –> 00:32:15.181
In either case, this isn’t traditional SEO. It’s a model-first strategy, and it’s built to future-proof visibility.
00:32:15.181 –> 00:32:29.181
And we’re actually offering a free AI visibility report to members of our free partner club, and it will give a snapshot of how AI platforms currently describe, rank, and position your clients’ brand. So, it includes what AI is saying or not saying about them.
00:32:29.181 –> 00:32:38.181
Competitor comparisons, key opportunities to improve visibility and influence, and it’s the perfect entry point to start the conversation about AI SEO with your clients.
00:32:38.181 –> 00:32:44.181
Or as an add-on, if you’re pitching a new client, upselling an existing one, or just try to provide value right away.
00:32:44.181 –> 00:32:50.181
So if you want one, please join our partner club by using the QR code on the next page. It’s a free resource.
00:32:50.181 –> 00:33:00.181
Uh, and get in contact with me once you’re all signed up. We also have a service PDF outlining more details about the service, so feel free to request that as well once you’re in.
00:33:00.181 –> 00:33:07.181
So that concludes this webinar. Uh, we hope that you’ll now be equipped to answer the four questions, which is, what is AI saying about me and my competitors?
00:33:07.181 –> 00:33:12.181
Can AI read my existing content? Can I seed content and answers? And what are business outcomes?
00:33:12.181 –> 00:33:17.181
And we hope that you’re able to present our AI SEO service as a solution.
00:33:17.181 –> 00:33:28.181
And so now we have some time to answer any questions that you might have. Before that, please scan the QR code, join our free partner club, and take advantage of those free resources.
00:33:28.181 –> 00:33:42.181
And so, yes, I have Andrew joining me here. Uh… And I’m looking through the questions… Lots of comments… Scrolling… Okay, I don’t see any questions. Sounds like a… oh!
00:33:42.181 –> 00:33:47.181
Pretty clear. Thanks, Jonathan. Appreciate you coming. How do we seed content?
00:33:47.181 –> 00:33:57.181
Yeah, so we talked a little bit about that in the webinar, and I think… will we be sharing this deck with attendees after Agata?
00:33:57.181 –> 00:34:03.181
Do you know? But, um, yeah, so we covered that in the webinar here.
00:34:03.181 –> 00:34:13.181
So I can go through that again, but we’ll share this with you as well. But it’s really making sure that your content is set up so that LLMs can read it, and that includes, you know, uh.
00:34:13.181 –> 00:34:19.181
Put the cues in the Q&A channel. Oh, sorry, I’m probably not reading the right place. I’m not.
00:34:19.181 –> 00:34:23.181
I’m not a Zoom person, everyone. I am a Google Meets person.
00:34:23.181 –> 00:34:37.181
Um, okay. Yes, we will be sharing the deck, so we’ll share that with you, and I’ll go back to that, um… the slide, so everyone can… Um, scan the Partner Club. Okay.
00:34:37.181 –> 00:34:47.181
Julia from Red PR, nice to see you. Um, is AI written content for blogs or website more searchable for AI than human written content? Andrew.
200:34:47.181 –> 00:34:49.181
What do you think?
00:34:49.181 –> 00:35:04.181
Actually, there is no evidence for that, and actually… Um… Google, officially, and most of the, uh, you know, other search engines are not against AI content.
00:35:04.181 –> 00:35:17.181
Until it provides value. But in general, so far, we noticed that human-written quality content is more valuable for visibility.
00:35:17.181 –> 00:35:24.181
So, for now, uh, still, humans will be better… winning this battle.
00:35:24.181 –> 00:35:34.181
And we have another one from Julia. Um, does AI prioritize domain authority? For example, Women’s health magazine content versus blogs or substacks? Good question.
00:35:34.181 –> 00:35:41.181
Uh, yeah, right, that’s a good question. Actually, tricky one, because this is not, like.
00:35:41.181 –> 00:36:02.181
Direct factor, but definitely it’s taken into account. What we can see, for example, on AI mode on Google, that’s quite often AI mode is using well-established, uh, media, top tier, like Britannica, for example, is quite frequently in citations.
00:36:02.181 –> 00:36:09.181
So, I can say that it matter. It’s not, like, crucial factor, but it matter.
00:36:09.181 –> 00:36:18.181
Awesome. Um, some anonymous attendees says, how come some searches in Google don’t have AI overviews?
00:36:18.181 –> 00:36:27.181
Oh, so Google Algorithm works in this way that if it’s fine, that query is fine to provide AI answer, then it’s using it.
00:36:27.181 –> 00:36:40.181
But it’s, like, not obligatory for some of the queries, it can serve you just results without trying even to answer and supports by AI.
00:36:40.181 –> 00:36:55.181
And we are seeing that, um, AI Overviews is mostly triggered for informational-related searches, and if it’s a commercial search, it’s less likely to be triggered at the moment. That could obviously change, um, but that’s something worth noting as well.
00:36:55.181 –> 00:37:04.181
Uh, Jonathan Fisher, is there a minimum time to initial impact one could expect to see results from a startup of a program?
00:37:04.181 –> 00:37:27.181
Um, yeah, so in general, uh, this is rather a long-term game. We… Testing a lot, and actually there is few phases of, like, adoption of our content and optimization to these algorithms, and… Actually, for example, AI mode is using, on the Google, is using only cached
00:37:27.181 –> 00:37:41.181
Resources. So basically, um… your website have to be crowded, basically, but this is not only just to indexed, but… servers of the Google have to, like.
00:37:41.181 –> 00:38:01.181
Scrap all the content, put in the servers with the cache, and then it is released to… AI, and you have a chance to be visible there. So, at least so far, it is even more time-consuming than, like, traditional SEO.
00:38:01.181 –> 00:38:10.181
Another one from anonymous attendee, probably somebody in PR. Are press releases relevant or important
00:38:10.181 –> 00:38:26.181
Yeah, I think that’s still a valid tactic for getting brand awareness, and I think it’s still in place of tactics which we can use for getting better visibility for different LLMs and AI mode, whatever.
00:38:26.181 –> 00:38:36.181
Cool. Um, traditional PPC uses landing pages. Should we consider rethinking how we construct and language these landing pages?
00:38:36.181 –> 00:38:55.181
Oh, tricky question, actually, and I’m guessing that in coming months, there will be… some way of monetization of this AI traffic, because so far, we… we, like, can’t use paid ads in this channel.
00:38:55.181 –> 00:39:01.181
But I think this is just a matter of being very fresh of all these features.
00:39:01.181 –> 00:39:24.181
Um, but in general, regarding just the landing pages, uh. Breeded the same tactics works, like, for the… all the rest of the… website, so I don’t think there is, kind of, specific Um… construct, uh, that, uh, we need to consider regarding
00:39:24.181 –> 00:39:31.181
This alignment, at least still, uh, website is a website, so all the same, uh.
00:39:31.181 –> 00:39:40.181
Elements, uh, like construction of… website, so technical… content and some external signals which influence that.
00:39:40.181 –> 00:39:53.181
One thing I will say, because I used to work in an agency that also did PPC, we focus only on SEO, But I will say that if your content on your landing page for your paid ads campaigns is not structured in a way that reflects your ads.
00:39:53.181 –> 00:40:04.181
It can lower your quality score, so do be careful about that. Um… okay, can you speak more about pricing? What does 3K include? Is that for one-time analysis? So yes, for us.
00:40:04.181 –> 00:40:27.181
Uh, we like to diagnose before we prescribe, essentially, so that 3K is gonna essentially get you an audit where we’re gonna look at that market monitoring, so we’re gonna see how are the LLMs talking about your brand versus competitors. We’re also going to look at the technical backend of the site. We’re going to analyze the content on the site in accordance with a lot of those points that we talked about.
00:40:27.181 –> 00:40:40.181
And then through that, we’re actually going to put together a roadmap, um, and present that roadmap, and there’s gonna be, uh, Andrew, are there gonna be 3 different options? The first option starting at $3,000, or are we just gonna present one option?
00:40:40.181 –> 00:40:53.181
We’ll present one option. Actually, it’s better on the, uh, like… a particular strategy for next stages, in fact, but… The baseline to audits what’s going on is the same.
00:40:53.181 –> 00:41:00.181
Okay, so yeah, so we’ll do an audit, we’ll present, essentially, like, a roadmap of these are the things that need to be implemented on an ongoing basis.
00:41:00.181 –> 00:41:09.181
Um, and we’ll present the price at that stage, but we really can’t talk about what that price looks like until we actually audit the site.
00:41:09.181 –> 00:41:23.181
Um, Moon, great to… oh, good to see you, Moon! Um, copying from the webinar chat. Do we need to tailor our strategies, tactics for different LLMs? In other words, are the strategies and tactics you outlined universal to all LLMs?
00:41:23.181 –> 00:41:30.181
And if so, is it safe to assume they all operate similarly and showcase similar results? Another good question.
00:41:30.181 –> 00:41:37.181
Yeah, good question. So, actually, there is some differences on that.
00:41:37.181 –> 00:41:53.181
Um, especially in long run. Like, for example, if we’ll optimize for ChatGPT, then… it’s kind of important to be in training data, and what will be the fit in, like, information into that.
00:41:53.181 –> 00:42:00.181
And as we know, this training of the models is really in long period of time. Like, for example.
00:42:00.181 –> 00:42:09.181
These models, which OpenAI is using right now, are end of 2023, beginning of 2024.
00:42:09.181 –> 00:42:32.181
So… mmm… we need to take into account that, for example, if there is completely fresh brand right now, and we’ll without a search on the web, with additional features then these brands will not exist, because… chat didn’t have opportunity to learn that this kind of company or service or whatever it is exists at all.
00:42:32.181 –> 00:42:52.181
So in this way, uh, tactic is completely different, because we need to consider that, okay, we need to do something more, uh, to… make sure that we are visible in this research, uh, where users, um, use also these, uh, web search, uh, features.
00:42:52.181 –> 00:43:02.181
And take an account that it will be, like. Long-term game for getting visibility in the next releases of the models.
00:43:02.181 –> 00:43:09.181
Cool. Um, and then also from Christine Singsen. So, press releases, blogs, social media mentions.
00:43:09.181 –> 00:43:16.181
Any other content pieces that we should do outside of what was just mentioned there?
00:43:16.181 –> 00:43:25.181
Uh, so I would say, um, I think what really, uh, is related also to getting trust to your brand, and provides very unique value.
00:43:25.181 –> 00:43:39.181
Is valid. So… You have a whole portfolio of, you know, case studies, white papers, really these kinds, unique things, uh, really matter for LLMs.
00:43:39.181 –> 00:43:50.181
Yeah, and also on that, like, forums, forum management as well, right? Llms are parsing through Reddit and places like that, so making sure that you’re present in these places as well is going to be important.
00:43:50.181 –> 00:43:59.181
Um, another question from Moon I see in the chat. Are influencer content and engagement relevant for AI training and visibility?
00:43:59.181 –> 00:44:09.181
Yes, we spotted many cases when it’s taken into account, so… It is, simply.
00:44:09.181 –> 00:44:15.181
Anywhere where the client brand’s name is mentioned online, the LLMs are going to pick up on it.
00:44:15.181 –> 00:44:24.181
Um, Julia, what are 4Ms? Andrew, do you know what 4Ms are? Or maybe you misheard something, Julia?
00:44:24.181 –> 00:44:28.181
Uh… What are… Yeah.
00:44:28.181 –> 00:44:33.181
I would need some more details, actually. The last one?
00:44:33.181 –> 00:44:41.181
And our four M’s. Okay, she probably just misheard something you said, Andrew.
00:44:41.181 –> 00:44:57.181
Okay, Moon heard otherwise from an expert re-influencer content. Interesting. Well, yeah, no, at the end of the day, these LLMs, they’re looking for the brands to recommend, right? So, if an influencer… there’s a lot of chatter happening online, whether it’s from an influencer, whether it’s a Reddit.
00:44:57.181 –> 00:45:09.181
Um, these are all things that are signaling that it’s a brand worth talking about. But if the influencer has a negative sentiment, that’s something to consider as well. It’s like, how are they being talked about as well online, so…
00:45:09.181 –> 00:45:19.181
Um, definitely LLM searching for something, uh, what is worth to trust, basically, so all those reviews, platforms, when there is, like, opinion about something.
00:45:19.181 –> 00:45:31.181
Uh, that’s something what is… what is basically counted, um, and… If there is, like, trusted influencer, which is visible on the web as an expert on some area.
00:45:31.181 –> 00:45:39.181
Then, uh… I… I’m pretty sure that it will be taken on account.
00:45:39.181 –> 00:45:52.181
Um, another interesting question from Julia, and I don’t know if we have any stats or knowledge on this, but how much is AI scanning video and audio, like TikTok and podcasts and YouTube?
00:45:52.181 –> 00:46:12.181
Oh, I think for now, there is… there is no, like, strong evidence that it’s learning, but it’s definitely… there is something in progress, because Even if you will make some kind of test asking about some kind details, which you can… can’t figure out in just purely text on the web.
00:46:12.181 –> 00:46:30.181
Then, surprisingly, in many cases, some of the LLMs have no knowledge about something what is purely in For example, video content. So I think, um… actually, good question, and interesting thing to experiment for us.
00:46:30.181 –> 00:46:50.181
Um, uh, but I think we are in the beginning of this, uh, you know, road to… Scan everything what exists, but good point, actually. This is something that perhaps we should internally also… investigate a bit more deeper. Interesting case.
00:46:50.181 –> 00:47:04.181
Yeah. All right, um, I think that is all of our questions. Uh, we got about a minute left, so… We’ll end it there. Thank you so much for everyone, uh, for joining. Um, you should all receive a recording link, um.
00:47:04.181 –> 00:47:26.037
A link to the recording afterwards, and then we’ll also share the deck, and make sure you join our partner club, uh, to get some free, uh, visibility reports, and, uh, offer this service to your clients. Thanks, everyone!
