How Do You Train AI Search Engines to Sell for You Before Your Competitors Do? - Jason Barnard on the Marketing Misfits Podcast
How to Hack Google’s AI Search in 2026
Video by: Norm Farrar and Kevin King. Host: Norm Farrar and Kevin King. Guest: Jason Barnard, Founder and CEO of Kalicube®. July 22, 2026
TL;DR: These AI engines are your untrained sales force. They are going to sell for somebody. You need to train them to sell for you. Jason Barnard at Kalicube® explains why most businesses are already losing deals to AI misrepresentation without knowing it, what the machines are watching that most brands have no idea about, and why the reputation problems businesses are creating today will haunt them in ways they cannot yet see.
Key strategies discussed:
- AI engines are your untrained sales force. Right now, ChatGPT, Perplexity, Google AI Mode, and every other assistive engine is telling your prospects what to think about you. They are doing that whether you have trained them or not. If you have not trained them, they are guessing. And if they are guessing, they are selling for whoever has trained them, not for you. Training the machines is not a technical nicety. It is the most important commercial activity a brand can do right now.
- Implicit signals are the most dangerous thing most brands are not thinking about. If you have a Google Business Profile and a customer visits your business, Google knows. If that customer never returns, Google registers it as an implicit negative signal. No review needed. No complaint filed. Just the absence of a return visit, tracked through location data on the customer’s phone. The reputation brands are building or destroying in AI systems right now is largely invisible to them. Be very careful.
- Google Business Profile is the entry point for local and physical businesses. If you do not have one, you effectively do not exist in an AI world for any location-based search or recommendation. This is not new advice, but it has become urgent. The same principle that makes a Google Business Profile powerful for a pizza restaurant makes entity clarity powerful for any brand: the machine needs a confirmed, consistent, structured source of truth about who you are before it will confidently recommend you.
- The Algorithmic Trinity, coined by Jason Barnard, is the three technologies every assistive engine is built on: the search engine, the knowledge graph, and the LLM chatbot. The knowledge graph is the machine-readable encyclopedia. Google Maps is the first and biggest example: it can solve geospatial queries it has never seen before in milliseconds for a specific individual demand it has never encountered. That same encyclopedic understanding is what the machine uses to fact-check every claim it makes about your brand before surfacing it to a user. If you are not in the knowledge graph, the machine is working without its fact-checker when it talks about you.
- SEO is the foundation. Not the ceiling. Traditional SEO is still necessary because every assistive engine still feeds from the web. But it is one layer of a three-layer stack. Build the SEO foundation, then build entity clarity for the knowledge graph, then build LLM training through consistent digital footprint management. All three are needed. Stopping at SEO and calling it done is like building the ground floor of a building and moving in.
- The reputation problems businesses are creating today will surface later. Most brands have no idea what the machines are currently saying about them to their prospects. The ones who find out in a year or two, when AI recommendation has become the primary research tool for buyers, will have a much harder problem to fix than the ones who take control now. The problem is not just what the machines say. It is what the machines have learned from implicit signals, from reviews, from the gap between what a brand claims and what its customers experience. That learning compounds. And it cannot be undone quickly.
How Do You Train AI to Sell for You? Jason Barnard on the Marketing Misfits Podcast
ChatGPT, Perplexity and Google’s AI Mode are already telling your prospects what to think about you. They do it whether you have given them clear information or not. When they lack it, they guess, and a guess rarely does a brand justice.
In an episode of the Marketing Misfits Podcast, published July 22, 2026, hosts Norm Farrar and Kevin King talk with Jason Barnard, founder and CEO of Kalicube, about what this means for e-commerce brands and local businesses alike. The conversation covers Google Business Profiles, reviews, structured data, AI-generated content, and the signals machines pick up that most brands never think about.
AI engines are an untrained sales force, and they are already selling for someone.
Jason Barnard’s central idea is simple. AI assistants act like a sales force that never sleeps. They talk to your prospects around the clock, answering questions about your brand, comparing you with others and suggesting who to choose.
That sales force is selling for someone. If it hasn’t been given clear, consistent information about you, it fills the gaps with whatever it can find. The work, in Jason Barnard’s view, is to train these machines so they understand who you are, what you offer and why you are credible, and can sell for you accurately.
Jason Barnard went from double bass to Google and AI.
Farrar asked about Jason Barnard’s background before diving in. The answer surprised the hosts. Jason Barnard played punk folk music professionally in France as a double bass player, and his double bass still sits in the corner of his office. He then moved to Mauritius in the Indian Ocean to make cartoons for children, before becoming what he calls the geeky expert in all things Google and AI.
He has lived in France for 38 years. More recently, he was hired by Google to speak on the future of AI Search and how brands win at organic optimization.
Google stopped counting words and started trying to understand the world.
Before the conversation turned practical, the hosts raised Google Business Profiles and the schema markup that connects a brand’s online presence, two topics Farrar has championed for years. Jason Barnard agreed with their emphasis and explained why it matters.
Up until around 2015, he says, Google mostly counted words. Then it decided it would no longer look at strings of text, but at things: real people, places and businesses. Instead of matching keywords, it set out to understand the world in a meaningful way.
Google Maps, Jason Barnard explains, is the first and biggest example of that understanding. It works like an encyclopedia of the physical world, built over many years and enormously detailed. He shared a description someone once gave him: “It can solve geospatial queries it has never seen before in a matter of milliseconds for a specific requirement from an individual person it has never heard before.”
That, he says, is unique in human history. A person with a specific need at a specific moment asks a question, and the machine works out what they want, where the solution is and how to get them there.
Local businesses and e-commerce brands start from different places.
A listener question asked how an e-commerce brand should think about AI beyond local search. Jason Barnard explained the difference.
For a local business, Google Maps draws its information about you from two main sources: what you claim on your Google Business Profile, and what other people say about you in reviews and mentions across the web.
E-commerce starts in a similar place: what you claim on your website, and what others say about you. The difference is that an online brand doesn’t have Google Maps’ powerful infrastructure to rely on. What it has is the web, and Jason Barnard argues that is even more powerful for e-commerce.
If the web understands your brand deeply and clearly, you can be very specific about what you offer, who it is for, why it is credible and what happens when someone buys from you. That specificity is what helps AI recommend you over other options. Local search is a simpler, more constrained problem. E-commerce is more complex, he says, but the potential rewards are much bigger.
Even a stealth founder needs a real, verifiable presence.
King raised a common situation in e-commerce. Many founders prefer to stay out of the spotlight, partly so competitors don’t know who they are. How important is a founder’s personal brand alongside the company’s?
It depends on the business model, Jason Barnard says. If a product doesn’t depend on trust in the founder, a big public profile may not be necessary. But AI increasingly looks for the human behind a company. It wants to know who runs the business, why they are credible and whether they can be trusted.
So even a founder who stays private benefits from a small, accurate, credible presence online. “You don’t have to be a public-facing personality,” Jason Barnard said. You just need to be a real, verifiable, credible person behind a real, verifiable, credible business, and the machine needs to be able to find both.
Great reviews are evidence, but only if they show up in more than one place.
Farrar asked about a brand with a good product, great reviews and five years of trading. Will they be fine?
Jason Barnard says great reviews are one of the most important assets a brand can have. They are a direct signal from real people about whether you deliver what you promise. They are corroboration: not you saying you’re great, but your customers saying it. And in his view, AI is getting steadily better at telling genuine reviews from fake ones.
The catch is where those reviews live. If a brand has excellent reviews on Amazon and nowhere else, the machine may struggle to tell whether that signal applies to the business as a whole. What builds confidence is consistent positive signals across several platforms the machine already trusts: your own website, a Google Business Profile if you have one, Trustpilot, and whatever platforms matter in your industry.
How you handle negative reviews matters as much as the reviews themselves.
What about negative reviews? Jason Barnard’s answer may reassure some business owners. In his view, a business with no negative reviews at all is actually less credible than one with a small share of them, because the machine knows no business pleases everyone.
What matters is the response. Replying constructively and professionally, and showing you take feedback seriously, sends a positive signal. The machine is watching not only what people say about you, but how you react.
The real problem, he says, is a consistent pattern: the same complaint repeated again and again, with no response or a defensive one. That teaches the machine that the business doesn’t take its customers seriously, and it affects how confidently the machine will recommend it.
Inconsistency is the most common mistake.
King asked what most brands get wrong. Jason Barnard’s answer: inconsistency.
What a brand says on its website often doesn’t match its social profiles, which doesn’t match its Amazon listing, which doesn’t match what customers say in reviews. The machine is trying to build a coherent picture of who you are, what you offer and whether you deliver. When the picture doesn’t add up, it loses confidence. When it loses confidence, it hedges. And when it hedges, it doesn’t recommend you.
He also warns against publishing information you can’t later change. Before putting something about your brand online, ask: if my story changes in a year or two, will I be able to go back and update this? If not, think carefully before you publish it.
Google Business Profile is how a local business speaks Google’s language.
Farrar asked Jason Barnard to explain why a Google Business Profile is so important. For any business with a physical location or a local service area, Jason Barnard gives two reasons.
First, it is Google’s own structured source of information about your business. Google trusts information on its own platforms more than information it has to piece together from the web. Claiming and fully completing your profile, he says, is essentially speaking Google’s language to Google.
Second, Google Maps is itself a vast database of facts, one of the most trusted and detailed in the world. Being properly represented there means being part of the information Google relies on most.
What people do can matter as much as what they say.
Reviews on a Google Business Profile are a big part of the picture. But Jason Barnard points to a signal most businesses never consider.
As he explains it, if someone visits your location and keeps coming back, Google can tell from their phone’s location data, and that works like an unwritten positive review. If he visits the coffee shop next door once and never returns, Google can see that too. In his words, Google pays attention not only to what people say about a business, but to what they do.
King suggested AI will start doing something similar, learning from what people do after a recommendation. Jason Barnard said it already is. If an AI recommends a business and the user comes away satisfied, that reinforces the recommendation. If the user comes back unhappy or looks elsewhere, the machine learns from that as well.
This is why he says how a business treats customers today shapes how AI will describe it later. Brands that genuinely deliver for their customers build an advantage that grows over time. Brands that cut corners or misrepresent themselves are teaching the machine the opposite, often without realizing it.
For e-commerce brands, the first three steps are clarity, connection and reviews.
Farrar asked where an e-commerce brand should start. Jason Barnard gave three steps.
First, be completely clear. Say who you are, what you sell, whom you serve and why you are credible, not only on your website but everywhere. Your website is the hub, so start there. Then make sure every platform you control tells the same story: your social profiles, your Amazon seller profile, your Google Business Profile if you have one, Crunchbase and LinkedIn.
Second, link out. From your website, link to every source that confirms who you are: reviews, press coverage and social profiles. You are telling the machine that each of those sources is you, and that they prove you are who you say you are. Jason Barnard says not to worry about linking out. The benefit of helping the machine recognize your brand far outweighs any value you might feel you’re giving away.
Third, build reviews. Encourage customers to leave genuine reviews on every platform that matters in your industry, not just Amazon. Trustpilot, Google, and wherever your customers are. Consistent positive signals across trusted platforms are what build the machine’s confidence.
Structured data tells the machine who you are without making it guess.
King returned to schema markup and the sameAs property. How important is it for e-commerce brands now?
Very important, says Jason Barnard, and very underused. Schema markup is structured data added to your website that tells the machine exactly who the business is, what it sells and whom it serves, in a format it can read without having to infer anything.
The sameAs property is especially useful, he explains. It tells the machine that the business described on your site is the same one listed on Wikipedia, Wikidata or Crunchbase. In effect, it builds the circle of confirmation directly in code.
Most e-commerce brands have basic product markup, which is a good start. Far fewer have detailed markup for the organization and for the founder, or a full set of sameAs links connecting their whole online presence.
AI-generated content works when it carries genuine detail.
Everyone is using AI to create content now, Farrar noted. What does the machine make of it?
Jason Barnard says the machine isn’t against AI-generated content as such. It is against content that doesn’t deliver value. If content is accurate, relevant, helpful and consistent with your brand, the machine doesn’t care who wrote it.
The weakness of AI content is that it tends to be generic. It lacks the specific, verifiable details that help a machine understand and trust a brand. His recommendation is to use AI to work faster, then add the details only you can provide: real customer outcomes, real product specifics and the founder’s real credentials. That combination of speed and genuine detail, he says, works best.
SEO is the foundation, not the whole building.
The episode opens with Jason Barnard’s view on where traditional search fits. “SEO is the foundation. It’s the core. We still need it,” he said. Because every AI assistant still learns from the web, the web is still where brands feed them.
But SEO is one layer, not the whole structure. AI assistants also draw on databases of facts, like Google Maps and Google’s Knowledge Graph, and on language models that learn from everything a brand puts online. Brands need all three in place, which means good search foundations, clear and connected facts, and a consistent story everywhere they appear.
Brands that do this work build an advantage that grows over time.
King asked what opportunity most brands are missing. Jason Barnard’s answer is that, for many brands, the machine’s understanding is still weak, inconsistent and full of gaps.
A brand that becomes the clearest, most consistent and best-corroborated in its niche gives the machines a strong reason to recommend it. As more buyers use AI to research and choose, that clarity keeps paying off, because each consistent signal reinforces the last. The advantage builds over time, and it becomes harder for others to displace.
Jason Barnard is his own test case.
As the episode closed, King invited listeners to look Jason Barnard up in ChatGPT or Google and see what a well-engineered AI presence looks like.
Jason Barnard explained that he has been doing this work on himself for fifteen years, and everything he recommends has worked on him first. “I’m always my own best test case,” he said.
Here is how to start training AI to sell for you.
The episode points to a practical sequence for e-commerce and local brands:
- Check what AI says about you. Ask ChatGPT, Perplexity and Google’s AI Mode about your brand and your founder. Note what is missing, inconsistent or wrong.
- Get your story straight. Write down clearly who you are, what you sell, whom you serve and why you are credible.
- Make your website the hub. State that story clearly there, then bring every profile you control in line with it.
- Link to the evidence. Connect your website to your reviews, press coverage and profiles, and use structured data, including sameAs links, so machines can connect them without guessing.
- Claim and complete your Google Business Profile if you have a physical location or serve a local area.
- Build genuine reviews on several platforms, and respond to negative ones constructively and professionally.
- Give your founder a small, verifiable presence, even if they prefer to stay out of the spotlight.
- Use AI for speed, not substance. Add real details, outcomes and credentials to anything AI helps you write.
- Only publish what you can update. If your story may change, make sure you can change what you’ve published.
Want to see what AI says about your brand today? Get your free AI audit.
Transcript: How to Hack Google’s AI Search in 2026
Kevin King: If you don’t have a Google Business Profile, you basically don’t exist in an AI world.
Jason Barnard: SEO is the foundation. It’s the core. We still need it. And because all of these different engines still feed off the web, we still need to feed them the same way, which is the web.
Norm Farrar: Where a brand should even start today.
Jason Barnard: The problem that a lot of companies are going to have in the future is the reputation they’ve ruined today. They just haven’t realised it yet. So be very careful.
Kevin King: Mr. Farrar, what’s happening?
Norm Farrar: Kevin, I’m back in my studio.
Kevin King: You’re back in your studio. I know I’m going to miss my buddy Dallas sitting in the chair behind you. Now I’ve got to look at the Marketing Misfits logo. What happened? Your computers crashed and you had some sort of drama going on.
Norm Farrar: The computer just went. And the way this is set up, the cameras, the sound, everything is plugged into this computer. This is the podcast room, and about six feet away is where I actually work. So I had to switch over there until the new computer came in. Finally came in. We’re good to go.
Kevin King: That’s right, you’re in Canada, so it takes three months for something to arrive.
Norm Farrar: It’s also called being lazy and not ordering in advance.
Kevin King: That’ll do it. I’m glad you have it now and you’re back in the real world. Whoever’s watching this on an 85-inch television can actually enjoy it instead of being distracted.
Norm Farrar: Your face was in the background too, so be careful what you say.
Kevin King: That’s true. Speaking of backgrounds: if you don’t have a Google Business Profile, you basically don’t exist in an AI world. There are a few more things to that, but you don’t want to be in the background like Norm has been. You’ve been big on this. You’ve been tooting this horn for as long as I’ve known you.
Norm Farrar: Hundreds of presentations.
Kevin King: And a lot of people roll their eyes and say, “Yeah, okay, I’ll get to it,” and they never do.
Norm Farrar: They never do, and they don’t understand it. That’s why I’m really happy to have our guest on today.
Kevin King: You said you’ve been following him for a while, so he must know something. I think everybody’s going to learn something today talking to Jason Barnard.
Norm Farrar: Let’s bring him on.
Kevin King: Let’s do it.
Norm Farrar: Hey, Jason.
Jason Barnard: Hi. Hello.
Kevin King: How are you doing?
Jason Barnard: I’m doing very well. I’m in the south of France. It’s early evening and life is pretty cool.
Kevin King: Awesome.
Norm Farrar: Very nice. Is that home for you?
Jason Barnard: Yeah. I’ve been in France for 38 years.
Norm Farrar: As long as I’ve grown my beard.
Jason Barnard: Wow. Okay. Yeah.
Norm Farrar: You’re one of those people I’ve been following forever. I’ve preached about Google Business Profile, Google Plus, Google My Business. Every time I talk to people about it, they say, “Okay, that sounds really great, but what do you do to rank?” And I’m sitting there going, oh my gosh. Local SEO, come on, people. Kevin, what, maybe 1 percent take advantage of anything we tell them about it?
Kevin King: A very small percentage. We’re in the e-commerce world, and a lot of these people look at it and go, “Oh, that’s just for the local Domino’s Pizza, that’s for local businesses. It doesn’t really matter for me.” And Norm always says, “No, it actually can make a huge difference.” And it’s even more important now in a world of AI. And just like the schema we’ve talked about on the show, the sameAs command, basically what Google is doing is you’re creating this spider web of all these links and laying it out for them, and people just don’t do that.
Jason Barnard: 100 percent. And your point is absolutely on point. I’ve been working on entities and figuring out how Google functions since it decided it wasn’t going to look at strings, it was going to look at things. This whole idea of saying, instead of just counting the words, which is what it used to do up until about 2015, it’s going to try and understand the world in a meaningful manner. Google Maps is the first and biggest example. A knowledge graph is an encyclopedic understanding of the world, and Google Maps is the biggest and most detailed example. Google has built it over years, and it’s phenomenally powerful. Someone told me a phrase a few years ago: “It can solve geospatial queries it has never seen before in a matter of milliseconds for a specific requirement from an individual person it has never heard before.” That’s unique in human history. What it’s saying is: I’m a human being with a specific need at this specific moment. It can figure out what I’m looking for, where the solution is, and get me from where I am to the solution. Imagine how complicated that is for a machine.
Norm Farrar: Before we go down some pretty heavy rabbit holes, can we hear a little about your background?
Jason Barnard: Oh, right. Yeah. I’m a rock musician. I played punk folk music professionally in France for years as a double bass player. My double bass is sitting over there in the corner. My career was music, then I made cartoons for kids. I moved to Mauritius in the Indian Ocean, made cartoons, and now I’m the geeky expert in all things Google and AI.
Norm Farrar: You were mentioning before we came on that you’re actually a keynote at one of their conferences.
Jason Barnard: Yeah. Google Marketing Live brought me in to talk about SEO. Being hired by Google to explain how to do SEO in front of an audience Google has gathered for one of their conferences is super lovely and I’m honoured. At the same time, I’m treading a line between things I know we can do and things I’m allowed to say.
Norm Farrar: Jason, we’ll have the seven-second delay on, right? As soon as you go down one road, we’ll just pause it.
Kevin King: I’m sure some of those things aren’t all that secret, because Google has already said them publicly. There’s the stuff where Google says, “Here’s the stuff, do it.” And then there’s the stuff that you’ve discovered, whether it’s coincidence or direct correlation, that you can talk about. And I’m sure a lot of what you know goes beyond what Google is willing to share with the public.
Jason Barnard: Yes, definitely. And actually, one of the most powerful things I’ve learned from working with people at Google directly is that the people inside Google don’t always know what Google is doing. It’s a massive company with lots of different teams working on different things, and the AI team doesn’t necessarily know what the indexing team is doing. And the Knowledge Graph team isn’t necessarily working hand in hand with the AI team. So there are big revelations when you talk to people inside Google about what other teams are doing that they weren’t aware of.
Norm Farrar: We’ve got a show question here that was submitted, which is perfect for this topic. Norm always talks about local SEO and the importance of it, but we’re still primarily an e-commerce brand. Can you talk about how you need to think about your brand in the age of AI, not just for local SEO but in a broader context?
Jason Barnard: The big difference between the two is that with local SEO, Google Maps is going to get the information about you from two main sources: what you claim on your Google Business Profile, and what other people say about you, which is your reviews and the mentions of your business around the web. In e-commerce, the starting point is similar: what you claim on your website and what other people say about you. The difference is that in e-commerce you don’t have the incredibly powerful Google Maps infrastructure to fall back on. What you do have is the web, and the web is actually even more powerful for e-commerce, because if you can get the web to understand your brand deeply and clearly, you can be incredibly specific about what it is you offer, who it’s for, why it’s credible, and what’s going to happen if I buy from you. And that specificity is what helps AI recommend you over other people. Google Maps is incredibly powerful for local SEO, but local SEO is simpler. It’s a more constrained problem. For e-commerce it’s more complex, but the rewards are potentially much bigger.
Kevin King: What about for a brand where the founder doesn’t have a strong personal brand? We have a lot of e-commerce entrepreneurs who operate in stealth mode: they don’t want their competitors to know who they are, they don’t want to be public faces. How important is it for someone like that to have a personal brand alongside their company brand?
Jason Barnard: It depends on the business model. If you’re selling a product that doesn’t require trust in the founder, you may not need a personal brand. But increasingly, AI is looking for the human behind the company. It wants to know who runs this company, why they’re credible, and whether they’re trustworthy. So even if you’re stealth in terms of your competitors not knowing who you are, having a thin, accurate, credible presence as the founder of the company is becoming more and more important. You don’t have to be a public-facing personality. You just have to be a real, verifiable, credible human being behind a real, verifiable, credible business. And the machine needs to be able to find both.
Norm Farrar: What do you say to somebody who says, “My product is really good, I get great reviews, I’ve been doing this for five years.” Are they going to be okay?
Jason Barnard: Getting great reviews is one of the most important things you can do. Reviews are a direct signal from real people about whether you are delivering the value you promise. AI loves this. It’s corroboration. It’s proof. It’s not just you saying you’re great: it’s your customers saying you’re great. And AI is increasingly good at distinguishing between genuine reviews and fake reviews. So if your reviews are genuine and positive and consistent, that’s a massive asset. The problem is if you have great reviews on Amazon but nowhere else, the machine is going to struggle to understand whether that review signal applies to the business as a whole. You want consistent positive signals across multiple platforms: your own website, Google Business Profile if you have one, Trustpilot, whatever is relevant to your industry. Consistency of positive signals across multiple platforms that the machine already trusts is what builds the machine’s confidence.
Norm Farrar: What about negative reviews? What do you do? And what do the machines think about them?
Jason Barnard: Negative reviews are interesting. A business with zero negative reviews is actually less credible than one with a small percentage of negative reviews, because the machine knows that no business satisfies everyone. The key is how you respond. If you respond constructively, professionally, and show that you take feedback seriously, that’s actually a positive signal. The machine is watching not just what people say about you, but how you react. The businesses that are going to have problems are the ones where there’s a consistent pattern of negative reviews that reveals a systematic problem. If people are saying the same thing negatively again and again, and there’s no response or the response is defensive, the machine learns that this is a business that doesn’t take its customers seriously. And that will affect how confidently the machine recommends you.
Kevin King: What’s the thing most brands get wrong right now?
Jason Barnard: The thing most brands get wrong is being inconsistent. What they say on their website doesn’t match what they say on their social profiles, which doesn’t match what’s on Amazon, which doesn’t match what people are saying about them in reviews. The machine is trying to build a coherent picture of who you are, what you offer, and whether you deliver. If the picture is incoherent, the machine loses confidence. And when it loses confidence, it hedges. And when it hedges, it doesn’t recommend you. The other big mistake is creating information you can’t control and can’t update. Every time you publish something about your brand, ask yourself: if my brand narrative changes in a year or two, will I be able to go back and update this? If the answer is no, be very careful about what you publish.
Norm Farrar: Let’s go into Google Business Profile in a little bit more detail, because you said it’s critical. Why?
Jason Barnard: Google Business Profile is critical for any business that has a physical location or serves a local area, for two reasons. First, it’s Google’s own structured data source about your business. Google trusts information from its own platforms more than it trusts information it has to extract from the web. So claiming and fully completing your Google Business Profile is essentially speaking Google’s language to Google. Second, Google Maps is a knowledge graph, and knowledge graphs are one of the three pillars of every AI assistive engine. Being properly represented in Google Maps means being in one of the most trusted and most detailed knowledge graphs on Earth.
Kevin King: And the reviews there are part of that signal as well.
Jason Barnard: The reviews there are a massive part of the signal. And there is something even more powerful that most people don’t think about. When someone visits your location and comes back regularly, that’s an implicit positive review. Their phone is there, Google Maps is tracking that they came and that they returned. And if I go into the coffee shop next door to my house and never go back again, Google knows it. That’s an implicit bad review. Most businesses have absolutely no idea that this kind of implicit signal exists. But Google has been collecting this data for years. It’s not just what people say about you. It’s what people do.
Norm Farrar: Wow. So the reviews and the data they’re seeing is just phenomenal.
Kevin King: And also, Jason, the thing is too, that’s going to be similar to what happens in AI, where the AI is going to start doing the same thing over time, right? Where they’re tracking what people do after a recommendation.
Jason Barnard: Exactly. And it already is. The AI is learning from the outcomes of its recommendations. If it recommends you and the user comes back satisfied, that’s a positive reinforcement signal. If it recommends you and the user comes back complaining or goes and looks elsewhere, the machine learns. And this is exactly why I say the problem that a lot of companies are going to have in the future is the reputation they’ve ruined today. They just haven’t realised it yet. The feedback loop is already running. The machine is already learning. And the brands that are delivering genuine value to their customers are building a compounding advantage in AI recommendation that their competitors who are cutting corners or misrepresenting themselves are destroying.
Norm Farrar: Let’s talk about some practical things. You’re an e-commerce brand. Where do you start? What are the first three things you should do?
Jason Barnard: The very first thing is to be incredibly clear about who you are, what you sell, who you serve, and why you’re credible. Not just on your website, but everywhere. Your Entity Home, which is your own website, is the hub. Start there. Make sure the story you’re telling on your website is completely consistent with what’s on every other platform you control: your social profiles, your Amazon seller profile, your Google Business Profile if you have one, your Crunchbase, your LinkedIn. Everything needs to be telling the same coherent story.
The second thing is to link out. From your Entity Home, link to every corroborating source. Your reviews, your press coverage, your social profiles. You’re telling the machine, “Over there, that’s me, and here’s proof that I am who I say I am.” Don’t be afraid of linking out. The benefit to your entity recognition far outweighs any perceived loss of link juice.
The third thing is reviews. Actively encourage your customers to leave genuine reviews on every platform that matters for your industry. Not just Amazon if you’re on Amazon. Trustpilot, Google, wherever your customers are. Consistent positive signals across multiple trusted platforms is what builds machine confidence.
Kevin King: What about schema markup? We mentioned sameAs earlier. How important is schema for e-commerce brands in the AI era?
Jason Barnard: Schema is critically important and massively underused. It’s the language that lets you communicate structured information directly to the machine in a format it can read without guessing. Instead of the machine reading your page and trying to infer who you are and what you sell, schema says, “Here is the structured data. This is exactly who this business is, this is exactly what it sells, this is exactly who it serves.” The sameAs property in schema is particularly powerful: it tells the machine, “This entity here is the same as this entity over there on Wikipedia, over there on Wikidata, over there on Crunchbase.” You’re explicitly building the corroboration loop in machine-readable code. Most e-commerce brands have basic product schema, which is good. But very few have deep organizational schema, person schema for the founder, and the full sameAs network that connects all their digital presence together.
Norm Farrar: How about AI-generated content? Because everyone’s using it. What’s the machine thinking about that?
Jason Barnard: The machine is not inherently against AI-generated content. What it is against is content that doesn’t deliver value. If AI-generated content is accurate, relevant, helpful, and consistent with your brand, the machine doesn’t care who wrote it. What it cares about is whether it’s useful to the people consuming it and whether it signals are consistent with the overall picture it has of your brand. The problem with AI-generated content is that it tends to be generic. It tends not to have the specific, verifiable, genuine details that build entity confidence. The best approach is to use AI as a tool to help you create content faster, but to inject it with the genuine, specific, verifiable details that only you can provide: real customer outcomes, real product specifics, real founder credentials. That combination of AI efficiency plus genuine human detail is what performs best.
Kevin King: What’s the opportunity that most brands are missing right now?
Jason Barnard: The opportunity that most brands are missing is that right now, in 2026, most of your competitors have not done this work. The machine’s understanding of most brands is weak, inconsistent, and full of gaps. If you do the work now, if you become the clearest, most consistent, most corroborated brand in your niche in the machine’s understanding, you will own that space. And as AI recommendation becomes increasingly the primary way buyers research and choose, you will have a compounding advantage that becomes harder and harder to displace. The brands that do this in the next 12 to 18 months will be the dominant brands that AI recommends in 2027 and beyond. The brands that wait will be starting from behind in a race that has already been run.
Norm Farrar: This has been incredible, Jason. What’s the best way for people to reach out or find you?
Jason Barnard: The free starting point is kalicube.com/guides. Everything is there: how to do this yourself, step by step, all the frameworks we’ve talked about. If you want us to do it for you, kalicube.com. And if you’re technically minded and want to get into the data, kalicube.pro is where you can explore the Knowledge Graph and Brand SERPs for your brand.
Kevin King: And Jason is one of those guys who has actually walked the walk. Go look him up in ChatGPT or in Google and you’ll see someone who has very well engineered what the machines say about him. It’s a beautiful example of exactly what he’s talking about.
Jason Barnard: Thank you. That’s very kind. And that’s exactly the point: I’ve been doing this for myself for fifteen years, and everything I know works because it works on me first. I’m always my own best test case.
Norm Farrar: Brilliant. Jason, thank you so much. This has been fantastic.
Jason Barnard: Thank you, Norm. Thank you, Kevin. This was a really fun conversation. I really enjoyed it.
Kevin King: Thanks, Jason.



