Data Over Narrative

One of the most dangerous things in business is a compelling story that the data does not support.

We are built for stories. A good narrative is persuasive, memorable, and satisfying. It connects events into a clean cause and effect that makes sense of the world. The problem is that a story can be compelling and wrong at the same time, and when it is, it leads confident people to confident mistakes.

I have watched this happen more times than I can count. A team builds a narrative about why something is working, or why a strategy will succeed. The story is coherent and everyone believes it. And then someone finally looks at the actual data, and the data does not support the story at all. The narrative was a construction. The evidence told a different tale, and the evidence was right.

This is why I lean on data over narrative. Not because numbers are perfect, but because they are a check on the stories we tell ourselves. A narrative will happily confirm what we want to believe. Data is less cooperative. It sits there being true whether or not it fits the story, which is exactly what makes it valuable. The data does not care what is convenient.

The danger of narrative grows with how good the storyteller is. A persuasive person can construct a story so compelling that it overrides everyone's judgment, including their own. The more articulate the narrative, the more important it is to ask the unglamorous question. What does the evidence actually show? Often the most confident story is the one that most needs checking, precisely because its confidence is doing the work that evidence should be doing.

This does not mean narrative is useless. Stories matter for communication, for motivation, for making sense of data once you have it. The problem is not telling stories. The problem is letting the story lead when it should be following the evidence. Build the narrative from the data, not the other way around. The story should be a description of what the evidence shows, not a substitute for looking.

I try to hold my own narratives loosely for this reason. The story I believe about why something is happening is a hypothesis, not a fact, until the data backs it. When the two conflict, the data wins, even when the story is more appealing. Especially then, because an appealing story that is wrong is the most expensive kind.

So combine judgment with evidence. Tell stories, but make them accountable to the data. And when a compelling narrative meets inconvenient numbers, trust the numbers.

The story is seductive. The data is true. Follow the true one.