When the gut and the data disagree

Every leadership team eventually arrives at the same argument, usually in a room that's gone quiet. The data points one way. Someone senior, with twenty years of scars, feels strongly the other way. The expensive question hangs in the air: do we trust the model or the veteran?

The lazy answers are everywhere. "Always follow the data" sounds rigorous and gets you killed by a clean dataset measuring the wrong thing. "Trust your gut" sounds bold and gets you killed by confidence that's really just habit. Both camps mistake their preferred error for wisdom.

Intuition is compressed experience. Bias is compressed comfort. They feel identical from the inside.

What a gut feeling actually is

Experienced intuition is real and valuable. It's pattern recognition built from thousands of cases the conscious mind can't enumerate. When a veteran says "this won't work," they're often running a simulation on data they absorbed years ago and can no longer cite. Dismissing that as "anecdote" wastes the most expensive asset in the room.

But the exact same sensation — certainty without articulable evidence — is also what bias feels like. The sunk cost you don't want to admit, the strategy you championed, the worldview you formed in a market that no longer exists: all of it produces a confident gut feeling indistinguishable, from the inside, from genuine expertise. That's the trap. You cannot tell them apart by how sure you feel.

What clean data hides

Data has the opposite failure mode. It's not that data lies — it's that it answers precisely the question you asked, which may not be the question that matters. A funnel metric can be immaculate and still measure the wrong moment. A satisfaction score can be statistically sound and miss the emotion that actually drives churn. The number is true; the inference is wrong. And because the data looks objective, nobody interrogates the gap between what was measured and what was meant.

The referee: four questions

When intuition and evidence collide, I don't pick a side. I run the disagreement through four questions that usually reveal which one is lying.

• What exactly did the data measure, and when? If the dataset is six months old in a fast-moving market, the veteran's live read may simply be more current. Recency beats rigour when the world has moved.
• What would change this person's mind? Ask the gut-feel holder what evidence would make them reverse. If the honest answer is "nothing," you're looking at bias, not judgment — real expertise can always name its own off-switch.
• Has this intuition been right before, in this specific situation? Pattern recognition only transfers when the patterns match. A veteran of one category trusting their gut in a genuinely new one is extrapolating, not remembering.
• Is the data measuring the decision, or something adjacent to it? Map the metric to the actual choice. If there's a gap, the intuition may be filling exactly that gap with hard-won sense.

Ask what would change their mind. If nothing would, it was never judgment.

The synthesis, not the verdict

The best outcome is rarely "data wins" or "gut wins." It's that the collision sharpens both. The veteran's discomfort tells you where to go look in the data you haven't cut yet. The data tells the veteran which part of their pattern no longer holds. I've built forecasting models adopted as global best practice, and the ones that worked were never the ones that overruled the commercial team's instinct — they were the ones that gave that instinct something precise to argue with.

When the gut and the data disagree, you haven't found a problem. You've found the most informative moment available to you — two different intelligences pointing at the same decision from different angles. The job isn't to silence one. It's to make them negotiate in the open, with a referee that both respect, until what's left is a decision sturdier than either could have produced alone.