Off Kilter 206: The Race for Marketing's Mind Part 1.
tl;dr: Epistemological sovereignty is the new competitive advantage.
AI is about to crystallize marketing's philosophical chaos into permanent code. While competitors fight for data sovereignty, the real battle is for epistemological sovereignty—owning your theory of marketing truth. The clock is ticking. We have roughly 18 months to encode our own worldview before someone else encodes it for us.
In 18 months, marketing will run on someone's theory of truth. The only question is whose.
Currently, AI is infiltrating every corner, including creative automation, synthetic research, customer analytics, and campaign optimization. Most CMOs see the opportunity for efficiency gains and dive in. But they're missing the real game being played.
Every AI system encodes a worldview. When we train it on our data, we're not just teaching it what we do—we're teaching it what we believe. About customers. About value. About what marketing is for. And once those beliefs become encoded, they become invisible, automatic, and nearly impossible to change.
The opportunity is to create epistemological sovereignty—to own our own theory of marketing truth, which is poised to become a powerful form of competitive advantage. But only if we recognize what's happening before it's too late.
Which neatly brings me to the question that sparked this whole thing in the first place.
"Why Do Marketers Keep Falling for Snake Oil?"
A client asked me this last week. Simple question. Devastating implications.
It's not because marketers are naive. I've never met a dumb CMO. And it’s not because they’re not commercially savvy. They usually know their businesses pretty well. Nope. It's because marketing lacks what every serious discipline takes for granted: a coherent theory of truth.
Accounting has double-entry bookkeeping. Science has the experimental method. Finance has NPV and ROI. These aren't just tools—they're epistemological spines that define how each field creates, verifies, and acts on knowledge.
Marketing?
We disagree on how marketing creates value.
We disagree on what constitutes proof.
We disagree on what's real, what's noise, or what's noise disguised as math, insight, or a creative breakthrough.
Shit, we can’t even agree on what a customer is.
The Seven-Headed Hydra That’s Eating Itself
This isn't just academic hand-wringing. It's daily dysfunction. Marketing today operates atop seven incompatible epistemologies, and the contradiction is killing us:
The Power Players:
Financial-Attribution (The Boardroom Winner): ROI is truth. If you can't measure it, it didn't happen. Currently dominates because it speaks CFO.
Engineering-Optimizing (The Rising Force): Test everything, optimize relentlessly. Winning because it produces clear data, even if it misses the big picture.
Platform-Convenient (The Hidden Ruler): Whatever's in the dashboard is real. Dominates by default because it's easiest.
The Struggling Middle:
Behavioral-Causal: Patterns reveal truth. Strong in theory, weak in boardroom translation.
Customer-Centric: Journey mapping uncovers need. Popular in PowerPoints, ignored in P&Ls.
The Marginalized Meaning-Makers:
Cultural-Interpretive: Brands are meaning systems. Right about value creation, terrible at proving it.
Narrative-Coherence: Story shapes perception. Essential for differentiation, dismissed as "fluffy."
How AI Will Turn Chaos Into Catastrophe
Here's what should terrify you: Humans are epistemological acrobats. We switch frameworks instinctively. Every day, we navigate these contradictions. Brand building in the morning, performance optimization in the afternoon. Long-term strategy with the CEO, short-term results with the CFO. It's exhausting and incoherent, sure, but it works, just about. Now, this chaos is on the verge of catastrophe because AI can't do that.
When AI encounters the kind of philosophical circus we deal with daily, it does what machines do—picks the clearest signal and runs with it. So, unless we do something to tell it otherwise, this will likely mean:
Finance logic wins (cleanest data, boardroom approved)
Vendor assumptions dominate (baked into your tools already)
Short-term devours long-term (easier to measure)
Chaos happens (AI chooses via random emergence)
The Vendor War You Don't See
While we've been busy comparing features, buying software, and assembling our tech stacks, every major platform has been playing a deeper game for control of our minds:
Google needs you to believe marketing equals capture. Their entire business model depends on you forgetting that great marketing doesn't just harvest demand, it must also create it. Every feature, every algorithm, every dashboard reinforces this worldview.
Meta pushes addiction-as-engagement. The algorithm knows best. Humans are predictable. Never mind that building a brand and building a dopamine loop are different things.
Salesforce sells the fantasy that perfect process equals perfect strategy. As if great generals won battles through better workflow management.
Adobe promises to systematize creativity by building an impressive infrastructure for scaling mediocrity.
These aren't neutral tools. They're epistemological Trojan Horses. And the scariest part? Most marketers are wheeling them through the gates with gratitude.
Close your eyes for a second and imagine where this goes next: You are about to outsource your thinking to the platforms you already use, deepening the conflicting truths these platforms already display, which will then require you to deploy even more AI to make sense of the incoherence that results. And, well… you can see where this goes.
Tech wins, everyone else loses.
The Finance Colonization Already Underway
Want to know why marketing keeps losing budget battles? It's not because finance has better arguments. It's because theirs are more coherent.
The CFO's model of marketing might be completely wrong—reducing everything to last-click attribution, ignoring brand value, missing compounding effects. But it's consistently wrong. Your chaos-patched model might be right, but it's incoherently right.
Guess which one the board trusts?
This is how marketing became a cost center instead of a growth driver. Not through malice, but through epistemological surrender. Forced to adopt finance's worldview because we lack one of our own.
The 18-Month Window Before Positions Harden
Currently, our philosophical chaos is mediated by humans. Messy but manageable. Once it's encoded in AI? Game over.
Consider what's already happening:
Marketers are drowning in “dumb data” creating decision incoherence
66% report struggling to demonstrate the impact of campaigns
74% of global marketers are not confident in their audience data
70% of marketing professionals have experienced burnout in the past 12 months
50% report experiencing emotional exhaustion
Imagine encoding this confusion into systems that will confidently make wrong decisions thousands of times per second. Forever.
Within 18 months, almost every major marketing organization will likely have embedded AI fairly deeply into its operations. At this point, or soon after, the epistemological positions will crystallize, and worldview(s) will become an integral part of the AI infrastructure. While changing vendors might be expensive, changing how your organization understands truth may be nearly impossible.
The Sovereignty Advantage
But here's where it gets truly fascinating. For the first time, incoherence is a choice.
Unlike every technology before it, AI is epistemologically programmable. We can encode OUR theory of marketing truth, not the vendors, not the CFOs, and not whatever random emergent nonsense the AI might come up with by itself. While competitors converge around vendor defaults or finance logic, we can build systems that embody our specific understanding of value creation.
Think about this:
A B2B SaaS company may choose an engineering-optimizing epistemology to reduce time-to-value.
A Challenger bank may choose a customer-centric epistemology to maximize differentiation from incumbents.
A luxury automotive brand may choose a cultural-interpretive epistemology because it’s not selling cars, it’s selling meaning.
The winner isn't who finds the "right" epistemology. It's going to be who consciously chooses a superior one for who they are, and the situation they are in, and then encodes it with clarity.
The Cost of Inaction
Do nothing, and here's your future:
Every competitor uses the same platforms, with the same thinking tools, with the same incompatibilities, converging on identical strategies
Marketing becomes pure arbitrage—deepest pockets win
Your team becomes little more than prompt engineers for someone else's worldview
Strategic differentiation dies at the algorithmic level
This isn't speculation. It's already happening in programmatic advertising, where everyone optimizes for the same signals and wonders why performance has been completely commoditized.
The Race Has Already Started
Epistemological sovereignty—owning your theory of marketing truth—is about to separate winners from losers. Not at the tactical level, but at the foundational level of how marketing understands itself.
The companies that win won't be those with the best AI. They'll be those who encode their own coherent worldview, while competitors remain philosophically fragmented, unconsciously adopting whatever theory comes bundled with their tech stack or whatever is dictated to them by finance.
We have 18 months. Maybe less. Maybe more. Our competitors are already moving, even if they don't fully understand what they're doing.
The question isn't whether we'll operate under someone's epistemology. We will. The question is whether it will be ours or someone else’s.
Choose wisely. Choose quickly. Because once this concrete sets, it’s going to be bloody hard to change.
Next week: The epistemological power play. How different organizations can choose, adapt, and encode a worldview that fits their business, before someone else’s version of “truth” gets hardcoded on their behalf.
“Google needs you to believe marketing equals capture. Their entire business model depends on you forgetting that great marketing doesn't just harvest demand, it must also create it. Every feature, every algorithm, every dashboard reinforces this worldview.” 👏👏👏👏
Just a few days ago I’ve stumbled upon Rory Sutherland with his brilliant points about marketing. And one recurrent thought he doubles down on is – "great marketing needs irrational exploration of ideas", which are unmeasurable in advance. Yet they can be evaluated objectively afterwards and that irrationality might also be part of competitive advantage in AI-dominated world, because AI can’t come up with anything conceptually new, that it wasn’t exposed to previously. Not the current LLM-based at least.