Maps, Territories, and the Stories We Tell About Data

The Argument

Alfred Korzybski's phrase "the map is not the territory" has become a cliché, but it's a cliché because it's true and we keep forgetting it. Every dataset, every chart, every dashboard is a map — a simplified representation of something vastly more complex. The danger isn't bad maps. It's forgetting that all maps are simplifications.

My Response

When the Map Replaces Reality

I work with data professionally, and I've watched organizations slowly replace reality with their metrics. A hospital measures "patient satisfaction scores" and optimizes for the score rather than for actual patient wellbeing. A school measures "test performance" and teaches to the test rather than teaching to learn. The metric becomes the goal, and the original purpose dissolves.

Charles Goodhart formalized this as a law: "When a measure becomes a target, it ceases to be a good measure." It's one of the most important sentences in social science, and it's violated every day in every organization I've ever worked with.

The Seduction of Dashboards

Modern data visualization makes this worse, not better. A beautiful dashboard feels like truth. The clean lines, the precise numbers, the real-time updates — they create an aesthetic of certainty that the underlying data doesn't warrant.

I've sat in meetings where executives made million-dollar decisions based on a dashboard metric that I knew was calculated from incomplete data with questionable assumptions. But the dashboard looked authoritative. The map was beautiful, even though the territory was obscured.

What Data Can and Cannot Tell You

Data tells you what happened. It can suggest what might happen. It cannot tell you what should happen. The gap between "is" and "ought" doesn't close just because you have more data. A dataset showing that customers click more on red buttons than blue ones tells you about clicks. It tells you nothing about whether optimizing for clicks is a good idea.

This is the fundamental epistemological limitation that data culture refuses to acknowledge. More data doesn't produce more wisdom. It produces more information, which is a completely different thing.

Wider Lens

Nassim Taleb's concept of "the narrative fallacy" is relevant here. Taleb argues that humans are hardwired to turn data into stories, and the stories always feel more real than the raw numbers. A chart going up is just numbers. A chart going up with a story about why it's going up is a decision-driver. The story does the work, not the data.

Cathy O'Neil's Weapons of Math Destruction shows how this plays out at scale. Algorithms — which are just maps codified in code — can encode and amplify bias, creating feedback loops that punish the already disadvantaged. The map doesn't just simplify the territory. It reshapes it.

Pushback

None of this means data is useless. The alternative to imperfect maps isn't no maps — it's chaos. Data-informed decisions are better than gut decisions in most domains. The point isn't to abandon quantification but to hold it with appropriate humility.

What Remains

Every time I look at a dashboard now, I ask: what is this not showing me? The answer is always more interesting than what's on the screen.