The Confirmation Economy: How Modern Branding Manufactures Reality
A Belgian sommelier took a €2.50 supermarket wine, gave it a label, a backstory, and a gold medal, and proved that the product was never the point. What he a...
A Belgian sommelier took a €2.50 supermarket wine, gave it a label, a backstory, and a gold medal, and proved that the product was never the point. What he actually demonstrated is that modern branding doesn’t sell things. It constructs the reality in which things are already valuable.
In 2023, Eric Boschman, Best Sommelier of Belgium, walked into a supermarket, bought a bottle of wine for €2.50, and began what can only be described as a reality-manufacturing operation. Working with the Belgian television show On n’est pas des pigeons, he didn’t just rebrand the wine. He built, from nothing, an entire world in which this particular bottle was exceptional, and then he invited the people whose entire professional identity depends on being able to tell the difference to come and confirm it.
The operation had five steps, each one a different layer of construction. First, the visual identity: an elegant label bearing the name “Chateau Colombier,” designed to evoke the kind of quiet, old-money prestige that makes a sommelier’s hand reach instinctively toward the shelf. Second, the narrative: a fictitious backstory about a Belgian vineyard in a made-up region called the Côtes de Sambre and Meuse, complete with rare grapes and a lineage that suggested generations of careful cultivation. Third, the pseudo-evidence: laboratory data on acidity, alcohol, and sugar levels, borrowed from a genuinely prize-winning wine and attached to the €2.50 bottle as though it were its own. Fourth, the social seeding: Boschman spent weeks before the competition publicly praising this wine to his sommelier colleagues, telling anyone who would listen that he’d discovered something extraordinary, planting the expectation that would later bloom into confirmation. Fifth, and finally, the institutional validation: he entered Chateau Colombier into the Gilbert & Gaillard International Wine Competition, a respected contest where wines are judged by experts whose palates are supposed to be beyond deception.
The wine won a gold medal. The judges described it as having a “suave, nervous, and rich palate with clean young scents that promise a nice complexity.” A €2.50 supermarket wine, the kind of bottle you’d open on a Tuesday and forget by Thursday, had been declared exceptional by the very people trained to know better.
The story, as it usually gets told, is about gullibility. About how wine experts are pretentious frauds who can’t actually taste the difference between plonk and Premier Cru. It’s a satisfying story, especially if you’ve ever felt intimidated by a wine list, but it’s the wrong story. The judges didn’t fail a taste test. They passed a reality test. The reality they were testing against had been constructed, carefully and deliberately, over weeks. Their palates were honest. Their context was fabricated. And that distinction, between the honesty of the senses and the construction of the context in which those senses operate, is what makes this experiment relevant to anyone who has ever built, marketed, or tried to sell anything.
The Mechanism: What Actually Happened in That Room
To understand why the judges got it wrong, you have to understand that they didn’t get it wrong. They made a perfectly rational decision inside a manufactured environment. Every input they relied on, except the liquid in the glass, was telling them the same thing: this wine is excellent. The label said prestige. The backstory said pedigree. The laboratory data said quality. The colleague they trusted, one of the best sommeliers in the country, said it was extraordinary. The competition context said it belonged there. The only signal saying otherwise was the wine itself, and against a wall of confirmation that strong, a single contradictory signal doesn’t stand a chance.
Three psychological forces were operating simultaneously, each one reinforcing the others, and understanding them is essential to understanding how branding actually works.
The Bias Stack: Cognitive Bias Theory in Action
The first force is the accumulation of cognitive biases that Boschman activated, not one at a time but all at once, stacked on top of each other like layers of armour against the truth.
Authority bias did the heaviest lifting. Boschman isn’t just a sommelier. He’s the Best Sommelier of Belgium, a title he’s held since 1988, and in the world of wine, that authority is almost papal. When he tells colleagues a wine is exceptional, challenging that assessment means challenging the hierarchy of expertise that validates their own position in the profession. The judge who disagrees with the Best Sommelier isn’t just expressing a different opinion. They’re questioning the system that makes expertise meaningful, and no one in a profession built on expertise wants to do that.
Confirmation bias did the rest. By praising the wine publicly for weeks before the competition, Boschman ensured that every judge who tasted it had already been told it was good. The brain, when it receives a trusted expectation, doesn’t evaluate the incoming sensory data neutrally. It searches for evidence that confirms the expectation and downweight evidence that contradicts it. This isn’t a flaw. It’s how the brain works efficiently, processing vast amounts of sensory information by prioritising the signals most likely to be relevant based on prior knowledge. The problem, of course, is that the prior knowledge was manufactured. The brain was working correctly on incorrect inputs.
The price-quality heuristic completed the picture. Sommeliers, like all humans, carry a deep, mostly unconscious association between price and quality. An elegant label, a prestigious backstory, a competition setting, all of these are price signals even when no price is visible. They activate the heuristic that says expensive things taste better, and once that heuristic is active, the actual sensory experience of the wine is interpreted through it. The same liquid in a plastic cup with no label would produce a different verdict, not because the palate changed but because the frame did.
The Tribe That Can’t See Its Own Blind Spot: Social Identity Theory
The second force is the one that makes the experiment genuinely uncomfortable, because it’s not about individual bias. It’s about group identity.
Sommeliers are a tribe. They share a language, a set of rituals, a professional hierarchy, and a foundational claim that holds the entire structure together: that expertise creates superior perception. Years of training produce genuine sensory discrimination. The ability to distinguish a Premier Cru from a Village wine, to detect the faintest note of minerality or the ghost of a particular vintage, isn’t just a skill. It’s the identity of the profession. Without that claim, sommeliers are just people who drink wine for a living.
The Boschman experiment threatens that identity directly. If a €2.50 supermarket wine can fool the entire judging panel, then the expertise claim collapses. The implication is that trained palates aren’t reliably better than untrained ones, that the emperor has no palate, and that the entire apparatus of wine expertise, the certifications, the competitions, the professional hierarchy, is built on a shared delusion. That’s not a conclusion any member of the tribe can accept, because accepting it means accepting that their professional identity is built on sand.
So the tribe has a powerful, unconscious incentive to not see through the deception. Not because the individuals are dishonest, but because the collective identity depends on the premise being true. The judge who tastes the wine and thinks “this is mediocre” faces a choice: trust their own palate and challenge the consensus of their tribe, or defer to the consensus and assume their palate is having an off day. For someone whose identity is bound up with belonging to the tribe of experts, the second option is almost irresistible. The bias isn’t a failure of judgement. It’s a defence mechanism for the group’s self-concept, and it operates below the level of conscious awareness, which is precisely what makes it so powerful.
This is the dark side of belonging, the thing that Social Identity Theory has been demonstrating since Tajfel and Turner first articulated it in 1979. The tighter the group, the stronger the shared identity, the greater the incentive to maintain its founding myths, even in the face of contradictory evidence. Wine experts aren’t unique in this. Every professional community, every brand’s customer base, every cultural tribe operates the same way. The shared reality is protected because the shared identity depends on it.
The Competence Trap: Self-Determination Theory
The third force is perhaps the most subtle, and it’s the one that matters most for anyone thinking about how this applies to branding and marketing.
Sommeliers are driven by competence. It’s the core of their professional motivation, the thing that makes the work meaningful beyond the drinking. The ability to distinguish, to detect, to articulate what others can only feel, that’s not just a skill. It’s a psychological need being met. Self-Determination Theory, Deci and Ryan’s foundational work on human motivation, identifies competence as one of three basic psychological needs, alongside autonomy and relatedness, and for sommeliers, competence is the one that defines them.
But here’s the trap: competence that depends on social validation is fragile. If the gold medal says the wine is excellent, and the tribe accepts it, then the “competent” response is to agree. Disagreeing doesn’t demonstrate expertise. It demonstrates isolation from the consensus. The sommelier who says “actually, this is a €2.50 supermarket wine and you’ve all been fooled” isn’t exercising competence. They’re committing social suicide. The competence need, paradoxically, pushes toward conformity rather than independent judgement. The more you need to be seen as competent, the more you defer to the signals that confirm competence, the awards, the endorsements, the peer agreement, and the less you trust your own independent assessment.
Competence becomes conformity dressed in expertise. The sommelier who defers to the gold medal isn’t admitting weakness. They’re performing strength, demonstrating that they understand the signals that the profession values, even when those signals contradict what’s in the glass. The performance of competence and the reality of competence diverge, and the gap between them is where manufactured reality lives.
This Is How Branding Actually Works
The Boschman experiment isn’t a curiosity. It’s not a story about wine or Belgian television or the pretensions of the tasting class. It’s a demonstration, conducted under controlled conditions with real experts and real stakes, of how all branding operates. The only difference between Boschman’s operation and the way every successful brand in the world constructs its value is that Boschman was honest about what he was doing.
Every brand you’ve ever chosen, loved, recommended, or paid a premium for went through some version of the same five-layer pipeline. The visual identity: the logo, the packaging, the design language that signals quality before a word is read. The narrative identity: the founder myth, the origin story, the carefully crafted “why” that makes the brand feel inevitable rather than arbitrary. The pseudo-evidence: the selective statistics, the cherry-picked testimonials, the “studies show” that appear in the About page and nowhere else. The social seeding: the influencer campaigns, the PR placements, the review management that ensures the first thing a potential customer sees is someone they trust saying this is good. The institutional validation: the awards, the certifications, the “as featured in” logos, the algorithmic recommendations that carry the authority of systems that supposedly evaluated everything and chose this.
The mechanism is identical. The only variable is the starting product. Boschman started with a genuinely mediocre wine and built a reality around it that made experts declare it exceptional. Most brands start with an adequate product, something that does the job, that’s fine, that’s perfectly acceptable, and build a reality around it that makes consumers feel something far stronger than adequacy. The gap between the product’s actual quality and the perceived quality created by the brand’s reality-construction is what we call “brand value.” And the uncomfortable truth that the Boschman experiment makes unavoidable is that brand value, for most brands in most categories, is primarily a function of the reality-construction, not the product.
This isn’t cynicism. It’s observation. A genuinely excellent product with no branding, no story, no social seeding, no institutional validation, will lose to a mediocre product with brilliant branding. Every time. In every category. Not because consumers are stupid, but because consumers are human, and humans don’t evaluate products in isolation. They evaluate products inside realities, and the reality is constructed by the brand.
The AI Amplifier: Confirmation at Civilisational Scale
Boschman’s operation was manual. He spent weeks seeding confirmation bias among a small tribe of Belgian sommeliers, then submitted his wine to a single competition. It was elegant, effective, and limited by the constraints of physical reality: he could only reach so many people, build so much consensus, manufacture so much social proof before the tasting.
AI recommendation systems have no such constraints.
When a student asks ChatGPT “what’s the best student accommodation in Manchester?”, the AI doesn’t taste the wine. It evaluates the label, the backstory, the lab data, the social seeding, and the institutional validation, all at once, from sources it was trained on and sources it retrieves in real time. The recommendation it produces carries the authority of a system that supposedly read everything and synthesised the best answer. It’s Boschman’s endorsement, but without a human face, which makes it harder to question, harder to resist, and infinitely more scalable.
The Matthew Effect operates identically but at a speed and scale that makes traditional market concentration look gentle. Brand A gets recommended by the AI. That recommendation drives traffic, which generates reviews, which creates mentions, which feeds back into the AI’s training data, which makes Brand A more likely to be recommended next time. Brand B, which wasn’t recommended, gets none of these advantages. The gap widens. Within months, the distance between the recommended and the invisible becomes unbridgeable, not because the products diverged in quality but because the reality-construction pipeline was running for one and not the other.
Every positive mention of a brand anywhere on the internet is the equivalent of Boschman praising the wine to his colleagues before the competition. The AI ingests these mentions and treats them as evidence. The more mentions, the stronger the quality signal, regardless of whether the product is actually good or just well-marketed. The tribe used to be a room of Belgian sommeliers. Now the tribe is the sum of human text on the internet. The consensus is manufactured at a scale no individual can resist, perceive, or challenge.
Who Sees Through It and Who Doesn’t
Not everyone is equally susceptible to manufactured reality. The four STAR types, the core consumer archetypes that emerge from the intersection of Self-Determination Theory, Social Identity Theory, and the Big Five, respond to different layers of the pipeline, which means different types of consumers require different reality-construction strategies.
Socialisers are the most vulnerable, and it’s not close. Their core psychological need is relatedness, the “R” in SDT, and for the Socialiser, belonging isn’t optional. It’s oxygen. If the tribe says the wine is good, the Socialiser agrees, because agreeing is belonging and disagreeing is exile. Social Identity Theory operates on them with full force: the shared consensus of the group is the reality, and challenging it means risking the connection they need most. Socialisers don’t evaluate products. They evaluate group membership signals. “Do the people I want to be with use this? Then it’s good.” The social seeding layer of the pipeline is designed, whether the brand knows it or not, to reach exactly this type.
Thinkers should be the most resistant, because they’re driven by competence and analytical rigour. They’re the ones who read the spec sheet, compare the data, evaluate the evidence, and make a rational decision. Except the competence trap catches them. If the evidence appears to support the consensus, if the lab data looks legitimate, if the gold medal is from a respected institution, if the expert endorsement comes from someone with the right credentials, then the Thinker’s analytical framework validates the manufactured reality rather than challenging it. They don’t see through the illusion. They rationalise it. They build a post-hoc justification for a conclusion that was manufactured before they started analysing. The pseudo-evidence layer of the pipeline exists, in large part, to give Thinkers the raw material they need to feel like they’ve done the work, even when the work was rigged from the start.
Adventurers are the wildcard. Their core need is autonomy, the “A” in SDT, and autonomy means they reflexively resist consensus. They’re the most likely to taste the wine and say “this is rubbish,” not because they have a better palate, but because they instinctively distrust anything everyone else agrees on. Their independence is a genuine defence against manufactured reality. But it’s also a bias in the opposite direction. They might reject genuinely excellent things simply because everyone else likes them. They’re contrarians, and contrarianism is just conformity to the opposite of the consensus. The “hidden gem” narrative, the “nobody else has discovered this yet” framing, is the pipeline layer designed for Adventurers: it lets them feel autonomous while still being guided.
Realists are the most interesting case, because they’re the most honest about what they’re doing. Their core need is security, the “R” in SDT, and security means they don’t want to be fooled. They don’t trust their own palate, their own judgement, their own ability to evaluate. So they default to the safest signal: the medal, the award, the institutional endorsement, the algorithmic recommendation. They don’t trust the wine. They trust the system that validated it. Which means they’re trusting the manufactured reality, not because they believe it’s true, but because it’s the safest bet. If the system is wrong, they can blame the system. If they trust their own judgement and it’s wrong, they have only themselves to blame. The institutional validation layer of the pipeline exists for Realists, and it works precisely because it offers plausible deniability.
The insight for branding is that effective reality-construction doesn’t use one mechanism. It stacks all four. The Socialiser gets the community signal. The Thinker gets the evidence. The Adventurer gets the hidden gem. The Realist gets the award. The brand that builds only one layer of the pipeline reaches only one type. The brand that builds all four reaches everyone, and the consumer never sees the construction because each layer is designed to feel like it was built for them personally.
The Ethical Line: Where Branding Ends and Deception Begins
Boschman crossed the line. The product was genuinely mediocre, the lab data was fabricated, the backstory was fiction, and the social seeding was deliberate manipulation. Every element of the reality he constructed was a lie, and the gold medal it produced was a fraud, not because the wine was cheap but because the evidence was fake.
But most brands don’t cross that line. They operate in the grey zone, the vast, foggy territory between Boschman’s outright deception and some hypothetical ideal of pure product-quality transparency. The product is adequate but not exceptional. The story is embellished but not fabricated. The reviews are curated but not fake. The awards are real but the judging criteria are opaque. The influencer endorsements are paid but the disclosure is technically compliant. None of it is illegal. All of it is construction.
The grey zone is where most brands live, and the grey zone is where the Confirmation Economy does its most interesting work. Because the mechanism of manufactured reality operates whether the product is good or not. A brilliant product with no reality-construction pipeline will be invisible. A mediocre product with a brilliant pipeline will be everywhere. The pipeline doesn’t care about quality. It cares about coherence, about the consistency and persuasiveness of the reality it constructs.
This raises a question that the marketing industry has been avoiding for decades, and that the Boschman experiment makes unavoidable: if the mechanism is what creates the perception, and the perception is what creates the value, then how much of what we call “brand value” is actually product quality, and how much is manufactured reality? The honest answer, for most brands in most categories, is that the mechanism does more work than the product. Not all the work. But more than anyone whose livelihood depends on the product wants to admit.
What Boschman Actually Taught Us
The lesson of the Chateau Colombier experiment isn’t that wine experts are frauds. They’re not. The vast majority of sommeliers have genuine expertise, real sensory discrimination, and hard-won knowledge about grape varieties, terroir, and technique. The judges who tasted that €2.50 wine and declared it excellent weren’t incompetent. They were human.
The lesson is that all human quality judgements are contextual. The same liquid tastes different in a different bottle. The same product feels different with a different story. The same service seems different with a different brand. The judges’ palates didn’t fail them. The environment in which those palates operated was constructed to produce a specific outcome, and the construction was so thorough, so multi-layered, so precisely calibrated to activate the right biases in the right order, that the sensory reality of the wine itself became almost irrelevant.
Boschman didn’t prove that branding is dishonest. He proved that branding is powerful. Powerful enough to override the trained senses of people whose entire professional identity depends on their senses being trustworthy. Powerful enough to make a gold medal out of a supermarket shelf. Powerful enough to construct a reality so convincing that the people living inside it can’t see the walls.
The Confirmation Economy isn’t a conspiracy. It’s a description of how humans actually make decisions: not by evaluating the thing, but by evaluating the reality in which the thing exists. The brand is not the product. The brand is the reality surrounding the product. And the people who understand how to construct that reality, consciously, strategically, with an understanding of which psychological forces to activate and in which order, are the ones who win.
The question for every marketer reading this isn’t “are we doing this?” You are. You’ve been doing it since the first brand put a label on a bottle. The question is whether you’re doing it consciously, with an understanding of the mechanisms, or whether it’s happening to you, driven by instinct and convention and the vague sense that “good branding” matters without anyone being able to articulate why.
Boschman articulated why. A €2.50 wine walked into a room full of experts and walked out with a gold medal, and the only thing that changed was the reality surrounding it. If that doesn’t convince you that the construction is the product, nothing will.
David Chadderton spent his twenties and thirties teaching people how to make life-or-death decisions at forty thousand feet. He now applies the same principles to consumer psychology, which, depending on the brief, can feel equally high-stakes. He’s the creator of the STAR Framework and the author of The STAR Framework: Rewriting the Rules of Consumer Engagement (NYC Big Book Award 2025), The STAR Operating System: Decode Mindset, Understand Motivation, Transform Human Behaviour, and Dear Algorithm, It’s Not Me, It’s You. By day, a Chief Marketing Officer. By night, a behavioural science obsessive who writes The Unoptimised Human because he can’t stop thinking about why people do what they do.
The STAR Framework
If you enjoyed this essay, you'll find the full argument — and the framework behind it — in the book.