Ask most marketers to defend their value and you'll get a familiar shape of answer: we ran this campaign, it cost this much, it generated that much return. It's the language of finance, borrowed because finance is the language the rest of the business already respects. Rory Sutherland thinks this is marketers making a category error about their own value — using the vocabulary of accounting to justify something that was never, underneath, an accounting function at all.1
The claim, stated as plainly as Sutherland states it: "the value of marketing is not what we do, it's how we think."2 A marketing function's real contribution isn't the specific ads it produces — it's the different lens it forces the rest of the organization to look through. Marketing, done well, is the discipline in the building that starts every question with "what does this look like from the customer's actual experience," and that habit of mind changes decisions well outside anything you'd normally call a marketing decision.
Sutherland's own words for this are unusually blunt for a professional context: "it gives the business an entirely new outlook on the world which then presents totally new and innovative ways of solving problems."3 He connects this directly to his recurring point that many organizational failures are really failures to combine engineering competence with psychological insight — the two disciplines rarely sit in the same room, and marketing, at its best, is the discipline that's supposed to be fluent in both, translating between them.4
Sutherland calls his own claim "a bit wanky" — his word, deliberately self-deprecating, because he knows how it sounds inside an organization that wants numbers, not philosophy.5 That discomfort is the whole point of the page: a genuinely philosophical justification for a business function is much harder to put in a slide deck than an ROI figure, which is exactly why marketers reach for the financial-return framing even when it undersells what they actually contribute. The honest, harder-to-defend claim is also, on Sutherland's account, the truer one.
You're a marketer preparing a budget justification for next year, and the instinct is to open with last year's campaign ROI. You try something different: you open with a decision the company made in a completely different department — engineering, ops, customer service — that changed for the better because someone in the room asked "what does this actually feel like for the customer," a question that only entered the room because the marketing function had spent a year training people to ask it. That's a harder story to tell in a spreadsheet, and it's a truer account of what the budget actually bought.
Later, in a cross-functional meeting about a technical decision that has nothing to do with advertising, you notice the engineers arguing purely from a specs-and-efficiency frame. You don't pitch a campaign. You ask the customer-experience question instead, the one your function exists to keep alive in the room. If the decision changes because of that one question, you've just demonstrated the exact value Sutherland is describing — value that never shows up as a campaign line item, because it was never a campaign.
The reason Sutherland's claim is hard to defend in a budget meeting isn't that it's vague — it's that a way of thinking, by its nature, shows up as an absence rather than a presence. You can't point to the specific decision marketing "caused" in engineering or ops, because the whole claim is that marketing changed the frame the decision got made inside, not that marketing made the decision itself. A changed frame is invisible in exactly the way a prevented mistake is invisible: you can't easily produce evidence for the bad decision that didn't happen because someone in the room had been trained to ask a different question. This is structurally the same measurement problem this batch's Eurostar case documents, one level more abstract — the value is real, the causal chain is real, and neither one leaves a data trail a spreadsheet can capture directly.4
You can still find evidence of this value, even without a metric for it — you just have to look in the right place. Listen for the specific phrase "what does this look like from the customer's side" showing up in meetings that have nothing to do with marketing — an engineering standup, a finance review, an ops postmortem. If that question is being asked reflexively, by people who don't report into marketing, in rooms marketing doesn't attend, that's the fingerprint Sutherland is describing: a discipline that trained an organization to ask a question on its own initiative. Its absence is just as diagnostic — an organization where customer-experience thinking only ever surfaces inside the marketing department's own meetings has a marketing function that's contained its value inside a department rather than spread it through the building.
Even marketers who privately agree with Sutherland keep reaching for ROI numbers, and the reason isn't confusion — it's that the philosophical case, however true, doesn't survive contact with a budget process built entirely around comparable, fundable line items. An honest philosophical justification competing against every other department's clean ROI number will lose the allocation fight nearly every time, regardless of which one is actually more valuable to the company, because the budget process itself only knows how to compare things it can put in the same units. Marketers reaching for ROI language aren't betraying the deeper truth Sutherland names — they're translating it into the only dialect the room is capable of funding, a compromise forced by the process, not a failure of conviction.
The stakes of misjudging marketing's value aren't abstract. A company that fully believes marketing is a tactical, campaign-executing function will organize accordingly — marketing sits inside a budget line, reports on campaign performance, and never gets pulled into decisions outside its formal remit. That company forfeits the exact cross-functional value Sutherland describes: the discipline that catches "engineering plus psychology" problems before they ship, because nobody with that specific lens was in the room when the decision got made. The cost of this mistake never appears on a P&L labeled "marketing was too narrowly scoped" — it appears, indistinguishably, as ordinary bad decisions made by well-meaning engineers and operators who simply never had anyone in the room asking the customer-experience question, because that person's job title kept them in a different building.
This is Sutherland's personal, decades-of-practice conviction, offered without a formal study — [POPULAR SOURCE], weight accordingly. The real tension: an organization that fully accepted "marketing is a philosophy, not a deliverable" would struggle to budget for it at all, since philosophy doesn't have a natural line item, which may be exactly why marketing departments keep reaching for ROI language even when they privately agree with Sutherland's framing — the philosophical case is true and also organizationally inconvenient.
This directly extends Misaligned Optimization Metrics (same batch): if marketing's real value is a way of seeing rather than a measurable output, then judging marketing purely by campaign ROI is a special case of the general Eurostar-metrics trap — using the legible, fundable number (ROI) to stand in for the illegible, unfunded thing that actually mattered (a changed organizational lens). The two pages diagnose the same structural blind spot from two different entry points.
Psychology — Attribute Substitution: Trust as Proxy for Competence (same batch). An organization asked to evaluate something genuinely hard to measure (whether marketing has improved how the whole company thinks) substitutes an easier question (did the last campaign's ROI look good) — the identical substitution mechanism documented at the individual level, now operating on how a whole function gets evaluated. The insight: this isn't a marketing-specific problem, it's what happens to any function whose real value is a way of thinking rather than a discrete output — the same substitution trap would await a research function, a design function, or a culture function evaluated purely on their most legible deliverable.
Business — Persuasion as the Greatest Skill. That page argues persuasion is a foundational, cross-functional capacity rather than a narrow sales tactic; this page makes the structurally identical argument about marketing — both resist being reduced to a tactical department and insist on being read as a whole-organization capability instead. Read together, they suggest a broader pattern: the most valuable business functions are consistently the hardest ones to keep contained inside a department org chart, and organizations that try to contain them anyway consistently undervalue them.
Sharpest implication: if a business capability's true value is a way of thinking rather than a set of deliverables, and organizations can only reward what they can measure, then the most valuable capabilities are structurally the most underfunded ones — not because anyone doubts their worth, but because worth and measurability have quietly become the same question when they were never supposed to be.
Generative questions: