The value finance can't see

A spreadsheet is a map. It is an extraordinarily useful map, and like every map it leaves things out. The danger is not the omissions themselves; it is that the omitted things slowly stop feeling real to the people reading the map.

A companion to why data-driven firms converge: where that piece argued the spreadsheet only ever contains the past, this one is about two more columns it does not have. It is also about two of Rory Sutherland's arguments that are really the same argument about that blind spot, one aimed at the org chart and one at the product.

The map is not the territory. The trouble starts when the people holding the map forget there was ever a difference.

Naming a function quarantines it

Peter Drucker argued that a business has only two functions, marketing and innovation, and that everything else is a cost. His sharper claim was that marketing is not a department at all: it is the whole enterprise seen from the customer's side.

Sutherland reanimates that with a twist about language. The mistake, he says, was calling marketing "marketing" and letting it harden into a department, because the instant a capability has a department, everyone else gets quiet permission to stop doing it. Not my job; there is a team for that.

Naming a function quarantines it.

Once quarantined, the function gets defined by what it produces rather than how it thinks. Marketing becomes the brochure department, the listings department, the show-home department. But the valuable thing was never the brochure. It was the habit of mind: treating perception as the actual product, knowing that how a thing is framed changes what it is worth. That habit is exactly what the rest of the org stops practising the moment a department is there to own it.

The crack in the argument, which Sutherland glides past, is that departments exist for a reason too. "Everyone owns it" curdles fast into nobody owning it, no budget line anyone will defend, no one accountable when it fails. So the honest target is not "abolish the department." It is: distribute the mindset, keep a centre of gravity. The way of thinking belongs everywhere; one place stays accountable for it.

You see the identical disease in security. The moment "security" becomes a department, the rest of the organisation offloads it, the security team handles that, instead of security being how everyone thinks when they design and ship. The fix is the same shape: the mindset distributed across everyone who makes decisions, with a centre of gravity that owns the standard. The real leverage was always upstream of the team that gets handed the problem, which is also why bolting on more tooling and playbooks downstream rarely moves it.

The reason we love them is that they're unnecessary

The second argument lands on the product. Sutherland's favourite example is the little foil on a San Pellegrino bottle: it does nothing functional. Finance looks at it and sees cost with no function, and concludes waste.

The customer looks at the same foil, sees the same cost with no function, and concludes that the maker can afford to be generous, is confident, takes care. The "no function" is load-bearing in both readings. They are looking at one fact and disagreeing only about what it means.

The reason the foil can mean anything at all is that a signal only carries information if it is expensive to fake. Necessary features cannot signal, because everyone has them; only the gratuitous can. This is costly signalling, and it is well-trodden ground:

• Amotz Zahavi's handicap principle in biology: the peacock's tail is credible precisely because it is a useless burden no weak bird could afford.
• Thorstein Veblen on conspicuous consumption: spending that signals because it is plainly wasteful.
• Michael Spence's Nobel work on market signalling under asymmetric information.

Finance hates these features because they are unnecessary. The customer values them for exactly the same reason.

Here is the crack, and it matters more than the insight. This logic licenses an infinite amount of expensive nonsense. Most gratuitous cost is just gratuitous cost, read by nobody, signalling nothing. "Finance hates it, therefore it is valuable" is precisely as lazy as "the spreadsheet says so, therefore it is true," merely inverted.

The hard, unglamorous skill, which Sutherland makes sound trivial, is telling apart the waste a customer reads as meaning from the waste that is only waste. Contempt for the spreadsheet is not the same as wisdom.

The naming point is this same idea aimed at language instead of objects. "Nuclear" borrowed its dread from the weapon; an extraordinarily clean, dense form of energy got named after a bomb. A name is a free, enormously consequential framing choice that the spreadsheet has no column for, and calling clean energy after a weapon was an unforced error. Whether "kitten power" is the fix is a separate question. The point is that the framing was a decision, an unpriced one, and unpriced decisions are the ones an organisation makes worst.

Findings, not noticings

Both arguments share one failure mode: treating the absence of a column as the absence of a thing. If responsibility is not on the org chart, it must be no one's. If value is not on the P&L (the running tally of money in against money out), it must be waste. The map has no cell for it, so the territory is assumed not to contain it.

The positive version of the lesson has a name Sutherland never stops pushing: Obvious Adams, a 1916 business parable by Robert Updegraff that David Ogilvy adored. It runs about sixty pages, and being public domain, it is free to read. Its hero's entire gift is seeing the thing too obvious for clever people to say out loud, then having the nerve to act on it while everyone else hunts for something more sophisticated.

That is the move no spreadsheet can make for you. Data is good at findings and terrible at noticings: the obvious never shows up as a result, because it was never a question anyone thought to ask.

The honest snag, which the cheerful parable hides, is the gap between obvious-in-retrospect and obvious-in-prospect. Once someone names the answer, the whole room slaps its forehead: of course. But "of course" is hindsight.

We celebrate the obvious ideas that worked and quietly file the obvious ideas that flopped under "naive," so "just do the obvious thing" smuggles in survivorship bias. The real skill is not worshipping the obvious; it is separating the obvious-and-right from the obvious-and-wrong before the result is in, which is exactly as hard as everything else here.

So the discipline cuts both ways. Do not mistake the map for the territory, and do not mistake your disdain for the map for insight either. The signal finance hates might be meaning or might be noise; the function with no department might be everyone's instinct or nobody's; the obvious idea might be genius or might be naive. The work is the discernment the spreadsheet was supposed to spare you, and never could.

There is a fitting coda. Each time Sutherland plugs that obscure 1916 pamphlet on a podcast, it climbs back near the top of Amazon. A book about the power of noticing the obvious, turned into a bestseller by a man noticing out loud that almost nobody reads it. That is not an aside to his argument; it is the argument, performed live. The map is indispensable. It is just not the territory.