Stop Calling It Data-Driven

Stop Calling It Data-Driven

The spreadsheet had fourteen tabs. The conclusion was unambiguous. The decision went the other way.

This happens in organizations every day, and most people in the room know it when it happens. What they cannot seem to stop doing is calling it something it isn't.

"Data-driven decision making" has become the corporate world's most comfortable lie — not because the data is absent, but because it is present and then quietly overruled. The analysis gets done. The findings get summarized into a deck. The deck gets presented to people with authority. And then those people make the decision they were going to make anyway, for reasons that have nothing to do with the fourteen tabs.

That is not data-driven. That is politics with a garnish.

The honest methodology, in most organizations, runs something like this: identify what outcome will generate the least friction among the most powerful stakeholders, then work backward to find data that does not actively contradict it. The spreadsheet is not the engine. It is the alibi. It provides the language of rigor without the inconvenience of being bound by it.

None of this is new. Institutions have always laundered preference through the vocabulary of objectivity — once it was expert consensus, before that it was tradition, now it is analytics. The specific jargon changes; the underlying move does not. Power decides, then reaches for the nearest credible-sounding justification.

The counterargument, and it deserves a fair hearing, is that data is never the only input. Leadership involves judgment, context, relationships, and risk tolerance that no spreadsheet fully captures. A decision that ignores the political landscape of an organization can be technically correct and practically catastrophic. Fine. Granted. But that argument is an argument for honest pluralism — for saying "the data points here, and we are choosing to go there for these other reasons." It is not an argument for calling the whole thing data-driven when the data was decorative.

The damage is not merely semantic. When organizations systematically claim evidence-based reasoning while practicing something else, they erode the credibility of actual analysis. Analysts learn, quickly, that rigor is theater. They start producing the conclusions that will survive the room rather than the conclusions the data supports. The spreadsheet still has fourteen tabs. But now the tabs are built backward from the answer, and everyone pretends not to notice.

Call it what it is. The decision was made to minimize conflict among people with power. That is a legitimate organizational strategy. It is not shameful. It is, in fact, sometimes wise. But it has a name, and that name is not data-driven.

Honesty about methodology is not a small thing. It is the difference between an organization that can learn from its decisions and one that can only rationalize them. The data will keep saying what it says. The question is whether anyone in the room is still listening.

--- The Marrow: Organizations that call political decisions "data-driven" don't just mislead — they corrupt the analytical process itself, training analysts to produce conclusions rather than findings.

Key Sources: No external sources cited in raw input; the fourteen-tab spreadsheet detail is preserved as illustrative specificity from the author's experience.

What I Shaped: Preserved the author's core frustration and the "garnish" metaphor, which was the sharpest thing in the draft. Expanded the argument beyond one meeting into a systemic critique of how institutions use data as alibi rather than input. Added the concession paragraph (judgment and context are real inputs) to give the argument credibility before sharpening the rebuttal.