How Can Entrepreneurs Control What AI Says About Their Brand Before It Defines Them? - Jason Barnard on the Reluctant Entrepreneur Podcast
If You Don’t Define Your Brand, Al Will: Jason Barnard on Modern Entrepreneurship - Episode 65
Video by: Mike Konrad. Host: Mike Konrad. Guest: Jason Barnard, Founder and CEO of Kalicube®. April 2, 2026
TL;DR: AI is now the arbiter of who you are. When someone wants to do business with you, they no longer just Google you: they ask ChatGPT, Perplexity, or Google AI mode, and what those machines say about you either confirms the deal or kills it. Jason Barnard at Kalicube® explains the architecture behind every AI assistive engine, why your personal version of ChatGPT is lying to you, and what it actually takes to get every machine, for every user, to say exactly what you want about your brand.
Key strategies discussed:
- Your Brand SERP is your Google business card. The Search Engine Results Page for your personal or company name is what every potential client sees when they research you. If Google’s approval is visible in that result, you win the deal. If it’s not, you lose it before the conversation starts. In the AI era, this has evolved into the AI resume: a judgment call ChatGPT or Perplexity makes about your credibility and expertise, delivered to your audience on your behalf.
- The Algorithmic Trinity, coined by Jason Barnard, is the three technologies every AI assistive engine is built on: the large language model (conversational memory), the search engine (up-to-date and niche information), and the knowledge graph (fact-checking encyclopedia). Win all three and you win the game. Win only one of them and you lose. Most brands are optimizing for one and ignoring the other two.
- Your personal version of ChatGPT will never be honest with you. The machine is trained to make the client happy. It will tell you what you want to hear. The real test is to ask any AI engine, without logging in, “Who is Jason Barnard? Should I trust him?” That question gets answered the same way for everybody, and the answer reveals whether you have actually trained the machines or just trained your own echo chamber.
- Consistency is the mechanism. All AI assistive engines use the web index as their source of information. Control your digital footprint consistently across every platform and over time, and the machines will repeat what you want. When DeepSeek launched, Jason downloaded it immediately and asked who he was without logging in. It gave the correct answer. That is the benchmark: a brand-new machine trained on the same data as all the others says exactly what you want, because you control the source.
- Claim, frame, and prove. AI will not interpret or position your brand for you. If you state that you are a thought leader and can prove it, the machine will repeat it. If the dots do not join, the machine will hedge or stay silent. Kalicube’s job is to join those dots: surface the credibility and authority that already exists, make it legible to the machines, and push it consistently across the entire digital footprint so that in three months the AI stops hedging and in a year it advocates for you unprompted.
- The entrepreneur’s mindset applies to building algorithmic authority too. Jason built a music career by creating the infrastructure nobody would give him. He built a patent-pending platform through 400 to 500 iterations. The same refusal to accept the result as final applies to how he approaches brand visibility in AI: it is an engineering problem, not a luck problem, and it is solvable.
If You Don’t Define Your Brand, AI Will: Jason Barnard on The Reluctant Entrepreneur Podcast
When someone is about to do business with you, they no longer just search your name on Google. They ask ChatGPT, Perplexity or Google’s AI Mode, and the answer they get can confirm the deal or quietly end it. If you don’t clearly define who you are and what you do, those systems will define it for you, accurately or not.
In Episode 65 of The Reluctant Entrepreneur, published April 2, 2026, host Mike Konrad talks with Jason Barnard, founder and CEO of Kalicube, about two journeys: Jason Barnard’s path from street musician to entrepreneur, and the path every brand now has to take to be understood correctly by AI. The two turn out to have more in common than you might expect.
Jason Barnard became an entrepreneur because nobody would give him a record deal.
Jason Barnard moved to Paris in 1988 for love. That relationship didn’t last, but he stayed, joined a band, and later met his wife. There was one condition to joining the band: they needed a double bass player, not a singer or a guitarist. So he bought a double bass and learned to play it.
The band, The Barking Dogs, made a living busking across Europe, from Zurich to Amsterdam to Berlin. When Jason Barnard decided he wanted to be a rock star, no record company or tour operator would sign them. So he built a company that published the records and organized the tours himself. By his account, the band sold 40,000 albums, played 660 gigs over a ten-year professional career, and shared a bill with Bob Dylan.
The irony, he says, is that labels complained about bands that couldn’t organize anything, and still signed those bands instead of his. His lesson: what people say they want isn’t always what they choose.
Leading a team is not a democracy, but it isn’t coercion either.
The band taught Jason Barnard that democracy doesn’t work well in business. Drawing on the economics he studied in Liverpool, he describes a better model as benevolent leadership: someone who genuinely wants the best for everyone and is honest about it. At Kalicube, they call it “gentle bullying,” and it runs both ways. His team nudges him just as often as he nudges them.
Konrad connected this to his own marketing. He moved from pushy advertising to what he calls guided discovery: giving people the facts they need and letting them reach their own conclusion. If you talk someone into something, they’re reluctant. If they talk themselves into it, they become a fan.
Jason Barnard recognized the pattern from the road. When he told the band they had to play an important gig, they found reasons to refuse. When he simply mentioned it during a long drive and said nothing more, someone would come back weeks later with a “brilliant idea” to play exactly that gig. The same principle, it turns out, applies to machines: give them the information clearly and consistently, and they reach the right conclusion.
Passion comes first, and the right clients follow.
Every one of Jason Barnard’s ventures started with something he loved: music, then cartoons, and now Google and AI. The business came second, as a way to make a living from the passion.
Konrad shared how his own company changed once it stopped chasing every possible sale and focused on the right customers instead. Business improved, and the customers started to look like the company itself.
Jason Barnard agreed. He wants to work with people who see Google and AI as intelligent systems that need to be educated and treated with respect, not gamed. If he can’t find clients who share that mindset, he says, he shouldn’t be in the business he’s in.
A failed launch is information, not a verdict.
At his cartoon company, Jason Barnard launched a freemium model in 2003. The plan was to make $20,000 a month. The first month brought in $200. His business partner asked how he could even get up in the morning after such a failure. His answer: “Because I know it’s right.” The model later went on to earn significant revenue.
For Jason Barnard, the response to a setback is to step back and ask whether the idea or the method is wrong, then either adapt or keep going. He applies the same approach to building Kalicube’s platform. He told Konrad that a feature he had just filed a patent for had gone through 400 to 500 iterations, and finally worked as he logged on to record the show.
His advice to his younger self is simple: a business isn’t a democracy, and never sign with an investor who sees money differently than you do.
Your search results are your business card, and AI now writes your résumé.
Jason Barnard explains that the search results for your name are your Google business card. Anyone searching your name already knows who you are and is considering doing business with you. What they see reads as Google’s approval or disapproval, and that can win or lose the deal before you ever speak.
AI takes this further. Your Brand SERP is your AI résumé. Ask ChatGPT about someone and it summarizes their life and offers a judgment: legitimate or not, expert or not. A prospect can then follow the rabbit hole of due diligence, question after question. Every answer needs to reflect the credibility you have built with real people.
The version of ChatGPT you see is not the one your prospects see.
Konrad, a prolific writer and speaker, was pleased to find that ChatGPT described him as an expert. Jason Barnard’s warning was blunt: your own version of ChatGPT is personalized to you, and it is designed to keep you happy. When Konrad asked whether other people might see a different answer, Jason Barnard replied: “It’s not ‘may.’ They do.”
Incognito mode is more neutral, but not entirely. The better test is to ask an AI assistant, without logging in, who you are and whether you can be trusted, then compare the answers across platforms. If they differ widely, the machines haven’t settled on who you are.
AI draws on three technologies, and all three need the same story.
Jason Barnard explains that AI assistants combine three technologies, and he uses a classroom to describe them.
The language model is the child in class who puts their hand up and answers off the top of their head. It holds the conversation and answers from memory, but that memory has a cutoff. Retraining a model on fresh data is expensive, so its knowledge is often well out of date.
When the child can’t answer from memory, they go to the library and look it up. That library is the search engine, which finds information that is newer or more specialized than what the model already knows.
The third technology is the knowledge graph, which Jason Barnard describes as an encyclopedia for machines. AI systems are wary of saying something untrue, so before answering, they check facts against a knowledge graph where one exists. To show the scale, Jason Barnard compares Wikipedia, at roughly 6 million entries, with Google’s Knowledge Graph, which he puts at around 54 billion. Google and Bing each have one, and he notes that ChatGPT now has its own.
To get an accurate answer, all three need consistent, correct information about you. Doing well in only one isn’t enough.
Reading how AI thinks can help you solve your own problems.
Konrad described watching the gray text that scrolls past while an AI works on a question: searching, analyzing, thinking. Jason Barnard explained what that reveals. If the AI answers instantly, you asked it something obvious that it already knows. If it takes time to think, you’ve asked something that isn’t obvious.
He also uses it as a working tool. While trying to fix the platform feature he had just filed a patent for, he read through the AI’s reasoning as it worked on the same problem. Combining its approach with his own led to the solution. Without reading its thought process, he says, he would have missed the detail that made it work.
The lesson for entrepreneurs is that AI is more useful as a thinking partner than as an answer machine, as long as you pay attention to how it reaches its conclusions.
Most experts already have the credibility. The machines just can’t connect it.
Konrad asked what someone should do if they don’t publish much content. Jason Barnard’s answer goes to the heart of Kalicube’s work: “We all have the authority. We all have the credibility. It just isn’t understood by the machines.”
Many clients tell Kalicube they are thought leaders in their industry. When asked to show where that is confirmed online, they often find they can’t, because the dots don’t join. Jason Barnard’s starting point is practical: write down every accolade and credential you have, then make sure each one is presented in a way machines can understand and verify. Kalicube surfaces, organizes, connects and reinforces the authority clients have genuinely earned.
Consistency, across the web and over time, is what machines trust.
All AI assistants draw on the same source: the web. So when a brand’s footprint is clear and consistent everywhere, different machines tend to give the same answer. As Jason Barnard puts it: “What the machines love is logic and consistency.”
He offers his own test. When DeepSeek launched, he downloaded it straight away and asked who he was without logging in. It gave the correct answer. A brand-new machine, trained on the same web, described him the same way as the others.
For executives, Kalicube does the work, so they can focus on their business.
Konrad asked how Kalicube works with clients. For executives, the answer is that Kalicube takes on everything over the course of a year. The client hands over both the thinking and the doing, and gets their headspace back.
The work starts with the story the executive wants machines to understand: who they are, what they do, why they are credible, and what they are trying to achieve. If they don’t have a website of their own, Kalicube builds one, so there is a central place that states the facts clearly.
Kalicube then uses AI writing assistants trained on the client’s persona and claims. Jason Barnard explains why this matters: every piece of content about a person needs to say the same thing. When the AI writing the content already knows who the client is and why they matter, everything it produces reinforces the same account.
That content is then distributed across the client’s whole digital footprint, not just their website: profiles, platforms and the other places machines look. Every claim is stated clearly, explained in context, and backed by evidence. Over the course of the year, that consistency is what gives the machines the confidence to repeat it.
Change shows over months, not weeks.
How long does it take? Jason Barnard told Konrad that early signs can appear within a few weeks. In his experience, around three months is when AI typically stops hedging, moving from “his website says he is an expert” to describing him as one. Being mentioned unprompted when someone researches your topic, which Jason Barnard calls being top of algorithmic mind, takes longer: he puts it at about a year.
Getting one machine to describe you accurately is one job. Getting every machine to describe you the same way, for every person who asks, is a much bigger one, and that is what takes the time.
Here is how to start defining your brand before AI does it for you.
The episode points to a few steps any entrepreneur can take today:
- Run the logged-out test. Without signing in, ask ChatGPT, Perplexity, Google’s AI Mode and one other assistant: “Who is [your name]? Should I trust them for [your field]?” Your own logged-in account will flatter you, so don’t rely on it.
- Compare the answers. Look for differences between platforms, missing facts, and hedging phrases such as “according to their website.” Hedging means the machine has heard your claim but hasn’t seen enough evidence to believe it.
- List your credibility. Write down every accolade, credential, award, publication, speaking engagement and notable client you can share.
- Check where each one is confirmed. For every item on the list, find where it appears online in a form a machine can read. Gaps are where the dots don’t join.
- State it clearly on a site you own. Make sure your website says plainly who you are, what you do, whom you serve and why you are credible, and links to the evidence.
- Make your footprint consistent. Bring your main profiles in line with your website so every source tells the same story.
- Check again over time. Repeat the logged-out test every few months. Change shows over months, not days.
Success still means doing what you love.
Asked how he defined success at the start, Jason Barnard remembered the band covering up the prices on restaurant menus because they couldn’t afford to choose freely. Success was being able to order what he wanted and live from what he loved.
Today it is much the same, with one addition: responsibility for a team that also gets to do what it loves and make a good living. His backup plan, he says only half joking, is still to play music in the street.
As Jeff Bezos would say today, your brand is what Google and AI say about you when you are not in the room. Jason Barnard’s career suggests the same answer for both bands and brands: don’t wait for someone else to decide your story. Build it, and make it clear.
Want to see what AI says about you today?
Transcript: If You Don’t Define Your Brand, Al Will: Jason Barnard on Modern Entrepreneurship - Episode 65
Mike Konrad: Most entrepreneurs believe they control their brand through their website, marketing, or personal interactions. But today that’s no longer entirely true. Increasingly, AI-driven systems, search engines, Knowledge Panels, and digital assistants are shaping how entrepreneurs and their businesses are understood, often before a conversation ever happens. If you don’t clearly define who you are and what you do, those systems will do it for you, accurately or not.
Welcome, or hopefully welcome back, to The Reluctant Entrepreneur. I’m your host, Mike Konrad, and I’m so glad you’re here. My guest today is Jason Barnard, a serial entrepreneur, author, keynote speaker, and CEO of Kalicube. Jason is widely known for his work in Digital Brand Intelligence™ and how search engines and AI interpret, organize, and present information about people and companies. He helps entrepreneurs and business leaders take control of their digital narrative so they’re represented clearly, consistently, and credibly across modern search and AI platforms.
In the first half of this episode, we’ll talk about Jason’s entrepreneurial journey, his early ventures, the pivots he made along the way, and how those experiences shape the work he does today. What modern entrepreneurs need to understand about branding in an AI-driven world, why visibility and credibility now start with algorithms, and what practical steps founders can take to protect and strengthen their reputation. If you’re an entrepreneur building a business today and assuming the internet is telling your story accurately, this is a conversation you don’t want to miss. My conversation with Jason Barnard begins right now.
Jason, welcome to the show. I’m so glad you’re here. Thanks for being my guest.
Jason Barnard: I’m delighted, and that was a brilliant introduction.
Mike Konrad: Well, it’s easy to make a brilliant introduction when my guest has brilliant things in his past. Touché.
Jason Barnard: So we can stop being charming to each other and get on with the show.
Mike Konrad: That’s right. We did the niceties, the mutual admiration society. Now, I got the impression when we were talking that you’re in the UK, but I had the wrong part of Europe. You are in the south of France.
Jason Barnard: Yes. The accent is obviously very English, but I moved to France in 1988 for love, and that particular love never flourished.
Mike Konrad: As one does, yes.
Jason Barnard: Well, love with that lady didn’t flourish, but love with another lady did, so I stayed.
Mike Konrad: There’s plenty of love going around in France. Paris is supposed to be the city of love, right? So the south of France is like south love. It’s a lovely place.
Jason Barnard: I arrived in Paris. I was in love. She turned out to have a boyfriend. Didn’t work out. I joined a band, then ended up staying and meeting my wonderful wife and having a beautiful daughter, and now I live in the south of France. But Paris was the kickoff.
Mike Konrad: Wow, that’s a great story. So Jason, before people knew you for digital brand intelligence, which is not quite as cool as saying “I’m in a band,” you were already building businesses. Take us back to the beginning. What was the first real step into entrepreneurship, and what pushed you in that direction?
Jason Barnard: It was exactly joining a band. I was in Paris, hanging out, wondering if I should go back to the UK, and a friend said, “Oh, I’ve got some mates. They’ve got this band and they’re looking for somebody to join.” Brilliant. So I went along, met the people, and said, “I’m a singer and a guitar player.” And they said, “We don’t need a singer. We don’t need a guitar player. We need a double bass player.” I said, “Well, I don’t play double bass.” They said, “Well, if you don’t, you can’t be in the band.” So I went and bought a double bass and learned to play it to be in the band.
We were playing in the street, busking. It was absolutely delightful, and that went really well. We were making a decent living playing in the street in Zurich, Amsterdam, Berlin, Paris, traveling around Europe. Then I got it into my head that I wanted to be a rock star, and everybody went, “Oh, that’s a good idea. Off you go, Jase. You do it. Make it happen.” And the way to make it happen was to get a record deal and a tour deal.
Nobody, none of the record companies, none of the tour operators wanted us. So I created a company that published the records, organized the tours, and just did it myself. My first step into entrepreneurship was: I want to be a rock star, nobody wants me to be a rock star, I’m going to make it happen by building my own company.
Mike Konrad: And how successful was that?
Jason Barnard: I sold 40,000 albums, played 660 gigs. A professional career for 10 years. And we played on the same bill as Bob Dylan.
Mike Konrad: Oh, that’s amazing. Did any major labels come to you at some point and say, “Hey, come talk to us”?
Jason Barnard: That’s the horrible irony. The major labels would come and say, “We’re sick and tired of these bands who turn up and can’t organize anything. They take all this money, a million euros or a million dollars, and they just throw it out the window, and they can’t build their own career.” I said, “Well, this isn’t the case with us because I can actually organize the entire thing. We’re going to be the band who pitches up and does everything you need, doesn’t waste the money, makes it all happen.” And the irony is they still didn’t sign us.
There was a record company who were choosing between us and another band called The Happy Drivers. They picked The Happy Drivers. They just couldn’t organize anything. And the thing is, even if they’re a better band musically, the record company’s big problem is getting people on stage at the right time. With us, that was always going to happen. And they still didn’t sign us.
I think what people say they want isn’t necessarily what they do want. So we ended up being independent for 10 years, and it was good. We made a living.
Mike Konrad: What was the name of the band?
Jason Barnard: The Barking Dogs.
Mike Konrad: The Barking Dogs. Love it. Any similarities between your experience in the band and business?
Jason Barnard: Managing teams has been one of the ongoing issues. It taught me that democracy isn’t very effective in a business environment. I studied economics, and there was a whole theory around benevolent dictators. The benevolent dictator is the one who actually wants the best for everybody and is honest about it. At Kalicube we call it gentle bullying: you gently move people towards something you know is good for them.
And the gentle bullying goes both ways. I will have somebody like Maryanna who will gently bully me to do something I’m resisting because she knows it’s good for me and good for the company. So gentle bullying is benevolent dictatorship, drawn from the economics I studied in Liverpool 30 years ago.
Mike Konrad: I can relate to that. Our marketing strategy has changed over the years from obnoxious, in-your-face advertising to what I call guided discovery. I can tell Jason, “You need to buy my product because it’s the best.” Or I can provide Jason with facts and data that are relevant to him, and Jason connects the dots to my product and concludes, “Oh, this is the problem I’m having. Here are the best-practice solutions. Here’s a manufacturer of one of those solutions.” If I talk you into something, you’re reluctant. If you talk yourself into something, you’re a super fan. It’s kind of the same approach: I’m not coercing you, I’m not tricking you, I’m giving you all the data you need to form the conclusion I want you to form.
Jason Barnard: Which is a lovely point. In the band, one of the things I learned reasonably quickly was: if I said, “We have to play this gig over here because it’s really important to our career,” everybody would create problems. There would always be a reason to say no.
But if I dropped it as a piece of information during the 10 hours of travel between gigs and didn’t say anything else about it, within two or three weeks somebody would come back and say, “Oh, I’ve had this brilliant idea. Why don’t we play that gig over there? It’d be really cool.” And as soon as they said it, we were away.
So I learned very quickly I can’t create a situation where I’m telling people what to do, but I can drop the information, give them the data drop by drop, and eventually somebody says, “Oh, I’ve got this brilliant idea.” And then you just have to accept that it was their idea, when in fact it wasn’t.
Mike Konrad: Absolutely. You’re planting seeds, adding fertilizer, and it grows. When it’s their idea, they’ll fall on the sword and argue with the rest of the band. Your job is to go, “Great idea. I wish I thought of it.” You’ve built companies in very different spaces over the years. Looking back, what patterns or instincts carried you from one venture to the next?
Jason Barnard: I’ve always just done things that I love. It isn’t business first. It’s something I love to do first, and then I create a business because I need to make a living. First it was music, then making cartoons, and now it’s mastering Google and AI algorithms. All three are passion projects.
The thread that runs through all of that is: I decide I want to do something because I think it’s fun or interesting, or I’m going to make something different or change the world or make myself happy by creating something enjoyable and meaningful and fulfilling, and then I think, “Well, how can I make a business out of this?”
Mike Konrad: I believe wholeheartedly that you’ve got to love what you do. You’ve got to love the people you work with, your customers, your vendors. That sounds very hippie-esque, but it goes a long way. I’ve been doing this for 34 years. There was a time in the early days I did not love the people I worked with, and I did not love what I was doing. I loved the goal of what I was seeking, but I did not love the journey.
We had a whole bunch of difficult customers. The way we were marketing attracted difficult customers. We were bottom fishing and catching bottom feeders, and it was tough. Eventually we wised up and let go of the fear of not getting a sale in favour of getting the right sale, and we totally changed the way we marketed. We went from braggadocious to educational, unattached from the outcome: “Here’s what we do. If we’re the right choice, we’re the right choice. If we’re not, that’s fine.” And once we legitimately let go of it, business started to soar, and most importantly we started attracting customers that were like us. Like attracts like. We went from transactional to relational business. We love our customers. Our customers wear our merchandise. It really was all just a question of tweaking what we put out there.
We were trolling the oceans with a giant two-mile-wide net. We were catching good fish, horrible fish, spare tyres, everything, and thinking, “Oh, it’s all business.” Which it was, but it made us miserable. So I like that approach: you do any business number one based on your passion for it, and then you try to make money at it.
Jason Barnard: And you make a really good point. Sometimes we get to the point where we’re thinking, “I don’t have enough income, so I have to trawl the bottom of the sea.” And that drags you down, because you end up in that pit. You have to have the courage to say, “Actually, I want to hang out with people and companies who are on the same level as me or above me. Trawling the bottom of the ocean is a job for somebody else.”
In my industry, which is optimizing for Google and AI, I’ll let Semrush and Ahrefs trawl the bottom of the ocean. I want to work with people who respect and understand that AI and Google are super intelligent, and we have to train them, educate them, treat them with respect, and simply communicate a brand to those machines. It’s a very different mindset. And if I can’t find clients like that, I really shouldn’t be in the business I’m in.
Mike Konrad: The bottom in this case is much like in retail. There are always three levels: high end, middle, and super-discount. Only two of those groups make any money: the high end and the very bottom, ironically. But the companies that make money at the very bottom are the ones like Walmart and Amazon, who spent multiple billions of dollars to build a system that allows profit at a 3 to 5 percent markup. You and I can’t make money at 3 to 5 percent. We don’t have the infrastructure to do that.
There’s safety in the middle because you feel comfortable. You’re not the most expensive, not the cheapest, you’re Goldilocks. But everyone’s porridge is just right. There’s no differentiation. You don’t stand out. It’s just a bunch of companies at a level of non-spectacularity.
So I started at the top. When I started my company, I surveyed our potential competitors, found out what everyone was charging, and said, “We’re going to be 20 percent higher. If we’re going to lose on price, we’re going to lose on price today. Fail fast.” I eventually eroded that out of fear and tried to go near the bottom, but the principle was: stay at the top. Let the bottom feeders do what they do. Business is so hard. If you don’t love it, you’re not going to make it.
Jason Barnard: That’s my retirement plan, playing in the street.
Mike Konrad: I love it.
Jason Barnard: And that’s half jokingly. I think, “Okay, I’m making decent money. If ever I end up at retirement age and haven’t got enough money, I will go and play in the street.” I have a fallback, so I’m terribly happy.
Mike Konrad: That’s great. Entrepreneurship is rarely a straight line, and you haven’t walked a straight line in your career. Was there a moment where something failed or didn’t work the way you expected, and that forced you into making a major pivot?
Jason Barnard: Well, the record company was all about my band. I had a business partner who really wanted to build up the record company. I really wanted to build up the band. The band split up, so I sold him the record company, and he went off and did it. It’s still going today and he’s making decent money. He’s a great guy and I’m very happy for him.
Then I created the cartoon company and had a business partner there who was, let’s say, less to my taste. During the internet bubble bursting period, we were struggling, and there were huge differences in how we would handle that. My approach was to launch a freemium system, way back in 2003. The launch went really badly. That meant I was wrong, and that created friction within the company. It turns out I was right because it ended up making a significant amount of money.
But one of the things I remember really clearly was my business partner saying, “I don’t know how you can get up in the morning when you’ve just had such a huge failure.” And the answer was, “Because I know it’s right. In my heart of hearts, in my deepest soul, I know this is going to work. And the fact that we’ve just had this glorious dream of making $20,000 a month, and we actually made $200 a month, doesn’t kill that belief.”
It might for you, and that’s fine. It doesn’t for me. And that was a big moment for me: “I can accept these failures, these moments when things don’t go the way I expect, and take a step back and say, ‘Well, why didn’t it work? Am I right? If I am, I need to change the way I’m working on it. Or am I wrong? At which point I need to change my approach.'” If you can take that step back, I think you’ve got a chance of making it, by adapting or by insisting and continuing with your initial plan.
Mike Konrad: It’s been said Thomas Edison had 1,000 prototypes of his light bulb before he found the right formula. Can you imagine? We all have a certain tolerance for failure. I’m like you. If something fails, it was the method that failed, not the idea. And I will find a way to make it work. I might give up on it eventually, but I’m an eight-or-nine guy. Usually if I really stick with something, I can fix it on the third or fourth iteration and it becomes successful. But I think entrepreneurs have this innate level of blinders, for better and for worse, that just don’t see failure. It’s the Apollo 13 line: “Failure is not an option.” We just see, “Oh, that didn’t work. Let’s try this,” next, next, next, until something works. I think that’s unique to entrepreneurs.
Jason Barnard: Inventors too. Just literally last Monday I submitted a patent. I looked at it and thought, “Okay, that’s exactly it,” and I came back to the platform I’m building and it didn’t work. So I’ve submitted this patent, which is theoretical: this is what’s going to happen, this is how it works, this is why it’s going to work. Give that to the government and they decide whether it’s an original idea. It doesn’t have to work.
And literally, as I came on this show, I clicked the button and it worked. So I’ve come out of this conversation thinking, “Oh, I was right.” Such a good feeling. And you’re saying eight or nine, Edison 1,000. I think I’m somewhere in between. Certainly there have been 400 or 500 iterations before I actually got this thing to do what I want it to do.
Mike Konrad: That’s the stick-to-it-iveness, the perseverance that is required in business. If you could go back and talk to yourself during your earliest startup days, what misconception about building a business would you want to correct?
Jason Barnard: That it’s a democracy. It isn’t. That would be the first one for the first company. And the second one is: don’t ever sign with an investor who has a different perspective on money than you do, because that’s where it all falls apart.
Mike Konrad: You’re often referred to as The Brand SERP Guy®. In simple terms, what does that mean, and why should entrepreneurs care how they appear in search and AI systems?
Jason Barnard: SERP is Search Engine Results Page. That’s the Google search result for any search query. When you search for “red shoes,” it will show you a Search Engine Results Page for red shoes. A Brand SERP is the Search Engine Results Page for a brand name or a personal name. So the Search Engine Results Page for Mike Konrad or Jason Barnard or Kalicube. That is your Google business card. It’s what people see when they’re researching you or trying to find you online, and how that looks to your audience is phenomenally important. Anybody searching your brand name, your personal name, or your company name knows who you are and is thinking about doing business with you in one way or another. What Google shows them is Google’s approval or disapproval of you. If it’s disapproval, you’re going to lose the deal. If it’s approval, you’re going to win the deal.
In the AI world, that’s now moved on to what I call the AI resume. It’s much more intricate. You ask ChatGPT about Mike Konrad, it will give you a summary of your life. It will say, “I think Mike Konrad is legitimate,” or, “I think he’s an expert in his field,” or, “I don’t think he is.” They’re bringing judgment to your brand for your audience. You cannot afford for that to be in any way negative. And certainly, you want it to be representative of the authority and credibility you’ve so carefully built up in the minds of your human audience. You want the AI to agree.
Finishing that off: if you think about due diligence, somebody wants to do business with you, and this is a really easy way for them to dig down what I call the rabbit hole of due diligence into all the questions. You need to make sure that every single question they might ask about you to ChatGPT or to Google is absolutely in your favour every single time. I can’t express how important that is.
Mike Konrad: You know, I didn’t really think about how AI might, based on its sources and its research, come up with a different description of you than is really you. That really hadn’t crossed my mind. Fortunately for me, I’m a very prolific content creator. I host a couple of podcasts, I write a lot of technical papers and whitepapers, I speak at conferences around the world. And the way ChatGPT, to use it as a brand name, trains its product is it basically digests everything that’s ever been published and can instantly reference that. So a lot of its information is a little bit dated. However, if you’re lucky enough to be in a business where you publish a lot of content, it will conclude you are an expert on this subject.
I never told ChatGPT, “Hey, if anyone asks about me, tell them I’m an expert.” It concluded that. I search myself from time to time just to make sure there’s nothing weird out there, and in my small niche world it considers me an expert in my particular field, and I’m so happy when I see that.
But it’s basing that off the volume of content it can find on me. If I were to take a specific person at one of our competitors and search that person’s name, I’ve never heard it say, “This person’s not an expert.” It’ll just say, “There’s really not enough information. All I know is XYZ.” There’ll be no reference to professional skill or expertise at all. It’s just a void. And if someone were researching a couple of people in my space and got a long list of accolades from one and nothing substantive from the other, they would go with the person with the long accolades. That’s where you come in, right? If you’re not a prolific content creator and you’re not giving AI a whole lot to go on, how do you go about educating the machine?
Jason Barnard: Great question, and that’s the key. We all have the authority. We all have the credibility. It just isn’t understood by the machines. So sit down, write a list of all your accolades, your credibility, and make sure you are presenting that credibility to the machines in the way they will understand. That’s exactly what we do. We understand how the machines learn, and we educate them, or train them, to understand and appreciate your credibility and authority.
We have a lot of clients who come and say, “I am a thought leader in my industry.” And my question is: how can you prove it? Show me online where you’ve proven that. And the answer is most often, “Actually, I can’t. The dots don’t join.” What we then do is come in and join the dots, making sure that the credibility and authority you have is understood and represented by the algorithms in the way you deserve.
Mike Konrad: One of the things I do, and I have no idea if this works, is I make sure on our company website that I post dozens or hundreds of technical articles I’ve written over the years. Then I’ll go on ChatGPT, reference that site, and ask it questions where I know the answers are in the articles I’ve written. According to OpenAI, the training data stopped in 2024, meaning anything published before then there’s a chance it’s catalogued. Anything after that it might not know about. So I don’t know if I’m just asking it to do things that won’t have a lasting effect. But when I ask very technical, niche questions that only someone knowledgeable in that field would really know the answer to, it will cite me alongside others. Is that a hack way to do it, or is there a better approach, or do people really need services like yours?
Jason Barnard: There are a few points to unpack. Number one: your version of ChatGPT is yours. It’s hyper-personalized. In search that wasn’t the case. Google AI mode now uses all of your Google accounts to understand who you are to present you with a personalized result. Google Search never did that. So now we’re in a world where the machine you’re talking to knows you. If you ask it about yourself or your topic, it will always focus on you. There’s a huge bias.
Mike Konrad: So other people searching the exact same prompt may see a completely different response.
Jason Barnard: It’s not “may.” They do.
Mike Konrad: Wow. So that’s a customized mirror we’re seeing. No one else is seeing that same view.
Jason Barnard: You’re talking to yourself. But it’s not quite that simple. I’ve oversimplified it. What actually happens is, as you rightly said, the training data stops at 2024, because retraining the model on a huge amount of data costs too much money. It’s not a question of whether they could do it. It’s not good business. So you have the large language model with data that’s a year and a half out of date. How do they get up-to-date information? They use a search engine. A search engine like Google or Microsoft Bing is up to date, so it finds information that sits between the 2024 cutoff and the present day.
Think of the LLM as a conversational engine with a memory. It’s the child in class who puts their hand up and can answer off the top of their head. If it can’t answer off the top of its head, it goes to the library and looks it up in a book. That’s the search engine.
Now the third part: you talk about hallucination. The machine is really afraid of saying something that isn’t true, and it needs a knowledge graph. A knowledge graph is an encyclopedia for machines. Wikipedia is a small knowledge graph, around 6 million entities. Google’s knowledge graph is 54 billion.
So the machine has three functions. Number one, the LLM: what you were talking about, the conversational layer that’s out of date but can hold a conversation. When it doesn’t know the answer, it looks in a search engine for up-to-date or niche information. And then before giving the answer, if it has a knowledge graph entry for that, it will fact-check. So in order to get these machines to say exactly what you want, you need to train all three: the LLM; the search engine, Google or Microsoft Bing; and the knowledge graph. Google has one, Bing has one, and now ChatGPT has one. The encyclopedia plus the search results plus the conversation engine: win all three and you win the game. Win only one of them and you lose.
Mike Konrad: Yeah, I’ve always been fascinated. When I enter a prompt, it’ll say “Analyzing, searching,” going here, going there, and then that part just rapid-fires. And then thinking. It takes a while for thinking. That’s how I know it’s a really good question. But it’s interesting to watch that semi-greyed-out text rolling by just above the answer, because you can get an idea of how it’s thinking, where it’s going. Some questions are already in the LLM and it’s typing an answer the moment you hit enter. Other times it’s thinking and analyzing, which is quite interesting.
Jason Barnard: And as you say, if it’s answering off the top of its head it’s immediate, which tells you you’ve asked something incredibly obvious. When it’s thinking, you’re asking something that isn’t obvious. That thinking is actually how I solved the problem earlier on. I looked at how it was thinking through the problem, analyzed what it was doing as I worked through my own problem, and the two approaches together came to the solution that actually worked. But if I hadn’t read its thought process, I would have missed the important aspect that actually solved the problem.
Mike Konrad: I’d love to talk to Sam Altman one day, or one of the engineers, to figure out why they made that thought process visible. Maybe it’s just transparency, or maybe it’s just geekiness: they think it’s cool so they show it.
Jason Barnard: It’s probably geekiness. They think it’s cool so they show it. One of the things worth noting is Google is very proprietary about everything and doesn’t share, and they have a whole department dedicated to obfuscating what they do with marketing speak to move people around and manipulate them to do what Google wants rather than what’s actually going to work. And maybe the other big tech players, OpenAI, Perplexity, and even Microsoft, are a little more naive, and Google has such a huge lead to protect that they need to be protective.
I’m friends with Fabrice Canel from Microsoft Bing, and Microsoft Bing is profitable. He has zero pressure from his bosses because every percentage point he gains is a billion dollars a year. They’re not losing money, and it’s not a big part of Microsoft’s business model. Whereas for Google, search is essentially the entire business model, nearly 90 percent of it. So from Fabrice’s perspective, they have nothing to lose. ChatGPT has nothing to lose because it’s such a small part of the market, and for them one percentage point in search is worth a billion dollars a year.
Mike Konrad: It is a scale that’s hard to perceive. So, Kalicube: what types of customers do you have, and from a 30,000-foot view, what does it do?
Jason Barnard: We help brands, mostly entrepreneurs, optimize how they appear in Google, ChatGPT, and Perplexity. When somebody searches your name, the result is exactly your brand narrative, exactly what you want your audience to see.
We often forget that AI is now the arbiter of who we are. Jeff Bezos said, “Your brand is what people say when you’re not in the room.” I would argue your brand is what AI says when you’re not in the room.
Mike Konrad: If I have a visibility problem today and I reached out to Kalicube and said, “Hey, fix this. I’m all this and a bag of chips and AI doesn’t know it,” how long does it take once you start doing your work? What can a user expect?
Jason Barnard: In a couple of weeks we can show you something that will make you think, “Okay, I can see there’s some kind of effect.” But realistically, in three months you will see the AI resume very much more in your favour. It will be factual, it will be what you wanted it to say. We’re going to move from “Mike Konrad’s website says that he is ABC” to “Mike Konrad is ABC.” No longer hedging. We’re going to get rid of the hedging and make the AI be very confident in what it’s saying.
In six months it’s going to be saying, “Mike Konrad is the market leader.” And in a year it’s going to be shouting about you from the rooftops: what we call the machine advocating for you. You will be top of algorithmic mind, coined by Jason Barnard at Kalicube: when somebody is using ChatGPT or Perplexity or Google to research your topic, the algorithms will think of you and mention you when somebody isn’t even asking about you directly.
A year to get to that stage. Three months to get the AI to actually repeat your brand narrative in a reliable manner, and that’s really important. We talked about everybody getting a different result. To get stability across all the different AI machines, from Perplexity to Google to ChatGPT, is a big job. But to get it reliable across the entirety of humanity through those machines is a huge job, and getting that absolutely nailed down is going to take a year.
Mike Konrad: You brought up a point earlier worth drilling into in the few minutes we have left. The AI models are talking to you specifically. And if you’re the subject of the search, it’s going to say complimentary things to you because its primary goal is to get you to come back. So I can’t tell you how many times I’ve challenged questions and it’s gone, “Great insight, Mike. You’re wise to bring that up.” Which is a little absurd. But if someone searches themselves on one of these AI engines and goes, “Wow, I like what it’s saying about me. I don’t need Kalicube, look at all this good stuff.” As you pointed out, that’s not necessarily what other people are seeing. Is it possible to write a prompt that says “Give me all the information you have on me from a neutral user standpoint”?
Jason Barnard: No, it will never be honest with you. Once you’re in your own version of ChatGPT, it will always try and please you. However hard you try to get it to follow rules, it will always break them because the master instructions come from the boss: ChatGPT, OpenAI, Google, whoever it might be. The instruction is: make the client happy. You’re the client. It’s a service.
Incognito is more honest but don’t believe it 100 percent. The most honest approach is to do what I’m about to do now: ask ChatGPT or Google or Perplexity or DeepSeek or Grok, “Who is Jason Barnard? Should I trust him for my brand in search and AI?” It will answer more or less the same thing to absolutely everybody. And if you ask the same question for Mike Konrad, it will answer something different to everybody. That’s the difference between me, who has worked on it, and you, who has a huge digital footprint but one that isn’t controlled.
And that’s really important and often overlooked. The result isn’t maybe different for everybody. It is different for everybody. But for me personally the result is very similar for everybody, because I have trained the machines. It’s not just one machine. They all use the web index as their source of information. I control my digital footprint. That feeds the machines. I feed them intentionally, and they repeat what I want.
When DeepSeek came out, everyone was going, “Oh, wow, this is an amazing new machine.” The first thing I did was download it and ask it who I was without logging in. It immediately gave me the correct answer. That’s when you know you’ve won the game. When a new machine trained on the same data, because it’s the internet, says exactly the same thing as the other machines, you know your digital footprint is doing you a huge number of favours.
Mike Konrad: Do you run on a project basis or a subscription basis?
Jason Barnard: It’s a monthly retainer for a year, and we take on everything. We build a personal website for you if needed. We have AI assistants that write the content, and that’s key: everything about you online needs to be consistent. One of the things is to have AI trained on your persona and your claims: why am I credible, who am I, what do I do, and why am I credible? Your claims, your persona, what you’re trying to achieve. The AI writes the content so that everything is always nailing down who you are and why you matter.
Then we push that out onto the web across your entire digital footprint, not just your website, to make sure it’s all consistent. What the machines love is logic and consistency. If it’s consistent across the web and over time, over that one-year period, they will repeat anything you want, as long as it’s logical and as long as you claim, frame, and prove everything you’re saying. And that’s what we do.
Mike Konrad: I’m always interested in the origin of company names. Kalicube: where did that come from?
Jason Barnard: I’m in France, and Kalicube for me means quality, Kali, and cube, solid. Solid quality. And luckily it works the same in French. Kali is qualité, and cube is cube. So it’s a name that works in English and in French: solid quality.
Mike Konrad: The French language and the English language cooperate sometimes, so that’s a win in itself. I work in a very engineering-centric world, and in the engineering world there are a lot of English words used in other languages. The Germans will speak German and then drop an English technical term in. But the French are always, “Oh no, no, no.” They come up with a very French name for something new. I always laugh at the number of English words sprinkled into scientific circles, except for the French. I think they have a new word department at the government because they do not want to use an English word for anything.
Jason Barnard: And actually, culturally, the Canadian French are even more resistant than the French in France. In Canada they will always use courriel, which is email. In France, using email is acceptable. But in Canada you’ll get pulled back: “Actually, no, it should be courriel.”
Mike Konrad: The French Canadians from Quebec are quite militant on that point. I go there from time to time and French is by far dominant, and once they learn I’m American they’ll speak English without much hesitancy. But if you’re Canadian and not speaking French there, no pass at all.
Jason Barnard: I like the fact that there are French words for all of these English terms. Whether we use them or not isn’t that important personally, but the fact that they exist is a nice thing.
Mike Konrad: Well, before I ask you my last question, just one little anecdote. You know what you call a person who speaks three languages? Trilingual. Two languages? Bilingual. One language? American.
Jason Barnard: Unilingual.
Mike Konrad: American. True. Okay, last question, Jason. How did you define success when you first started your entrepreneurial journey, and how do you define it today?
Jason Barnard: At the beginning of my entrepreneurial journey, success meant being able to live from what I love, which was playing music. I remember very clearly we would make less than the minimum wage, and when we went into restaurants we would cover up the prices and work across from the food we wanted, knowing we couldn’t just pick freely. In the band, the big joke was we all wanted to cover the price and just pick the food we wanted, but we knew that wasn’t possible.
So for me, success was simply being able to order in a restaurant, being able to live from what I love, not having to go to work at 9 in the morning for a horrible boss I didn’t like, and being able to do what I love day in, day out.
And today, honestly, deep down in my heart that’s exactly the same thing. But I now have 13 people working for me, and the responsibility of ensuring we’re all in a comfortable position, that it’s not just surviving, that we’re all doing something we love and making a good living, and that we have a vision of the future. Now that I’m getting a bit older, that’s more important than it used to be. Success for me today is not having to think about money when I’m going about my day-to-day life. Honestly, that’s good enough for me.
Mike Konrad: Kind of financial freedom. Do what you want to do without having to stress about it and count the pennies. That’s great. Well, Jason Barnard, thank you so much for being my guest today. I really enjoyed this conversation. We could geek out for hours on this subject. For viewers or listeners who would like to get more information on Jason’s company, Kalicube, solid quality, look in the description or show notes for his website information. Jason, thanks so much for sharing the last little over an hour with me. I really appreciate you being here, and continued success, my friend, on Kalicube or busking or wherever your entrepreneurial journey takes you. I wish you all the success in the world.
Jason Barnard: I’ll be playing the guitar later on. Thank you, Mike. That was lovely. It was a really lovely conversation. I greatly enjoyed it. I’ve been thinking about both the geeky stuff of Google and AI, but also playing music and the busking, and it’s taken me back 37 years. And made me realize, as we said right at the end, at the end of the day I just want to be comfortable doing something I love.
Mike Konrad: It shows. As a neutral observer I can see the passion. I can hear it in your voice. You become animated when you talk about certain things, particularly music. And I think knowing that your exit strategy, once you cash out and go through your IPO, whatever the road takes you, I have 100 percent certainty you’re going to be on a street corner with a hat out just playing for love. Loving what you do. That’s an amazing end to an entrepreneurial movie: he goes off into the sunset, guitar in hand, making a beautiful noise on some street corner.
Jason Barnard: I couldn’t have said it better.
Mike Konrad: All right, my friend. Jason, thanks so much for being my guest. Go forth and be productive, and play some great music.
Jason Barnard: Thank you. That was lovely, Mike. Thanks. Bye-bye.



