What the data actually shows
Daniel Kahneman's framework distinguishes two modes of thinking: System 1, which is fast, automatic, and intuitive, and System 2, which is slow, effortful, and deliberate. Much of Kahneman's work catalogues how System 1 intuition produces predictable errors — overconfidence, neglect of base rates, sensitivity to how a question is framed — particularly in statistical or unfamiliar problems where the quick answer feels right but is systematically off.
Gary Klein's naturalistic research points the other way, and it is not a contradiction. Studying firefighters, nurses, and other experts making real decisions under pressure, Klein found that skilled professionals often make excellent, rapid choices by intuition — recognising patterns built from years of experience. That intuition is real expertise, compressed: the expert is not guessing, they are recognising a situation they have effectively seen before.
Kahneman and Klein resolved the apparent conflict together, identifying the conditions under which intuitive expertise can be trusted. Their joint conclusion is that valid intuition develops only in environments that are regular enough to be predictable and that provide adequate, timely feedback so the person can actually learn. Where those conditions hold — chess, firefighting, reading familiar people — intuition is reliable. Where they do not — predicting stock markets, long-range forecasting, one-off major life choices — confident gut feelings are not trustworthy, even when they feel just as certain.
Why this feels different from how it actually is
The trouble is that intuition feels equally confident whether or not the conditions for valid intuition are met. A gut feeling carries the same sense of certainty in a domain you have mastered and in one you have barely encountered, because the feeling of knowing is generated separately from whether you actually know. This is why 'trust your gut' is such seductive and such unreliable advice on its own.
Culture also pushes a clean, one-sided story. We celebrate the visionary who 'just knew,' and we rarely hear about the equally confident gut calls that failed, because survivorship makes good intuition look more reliable than it is. The result is a folk belief that trusting your gut is generally wise, when the evidence says it is wise only under specific, identifiable conditions.
At the same time, deliberation has a bad reputation it partly deserves: overthinking small, reversible choices is genuinely draining and often pointless. So people swing between 'just go with your gut' and 'analyse everything,' when the research actually supports neither blanket rule. The mode should track the decision, and the decision type is usually knowable in advance.
What the research says to do about it
Match the method to the situation. Before deciding how to decide, ask two questions the Kahneman–Klein work makes central: do I have real, repeated experience in this exact kind of situation, and has that experience come with quick, honest feedback? If yes to both, your intuition is likely trustworthy. If no to either — it is novel, rare, or you never find out whether you were right — lean on deliberate analysis instead.
Weight reversibility and stakes. The cost of overthinking a small, reversible choice is mostly wasted energy, so for low-stakes decisions a fast gut call is usually fine and frees you up. Big, irreversible, or expensive decisions justify the effort of slowing down: laying out the options, the probabilities, and the likely outcomes, precisely because intuition is least reliable in exactly these rarer, higher-stakes situations.
Where you can, combine the two rather than choosing. A useful pattern supported by the decision research is to do the deliberate analysis and then check it against your gut: if your intuition strongly disagrees with the analysis, that disagreement is a signal worth examining — sometimes your experience is picking up something the analysis missed, and sometimes a bias is. Treating them as cross-checks tends to beat trusting either alone.
What the research says does not help
Adopting 'always trust your gut' as a life rule is not supported by the evidence and is actively risky in unfamiliar or statistical decisions, where confident intuition is most likely to be wrong. The strength of a gut feeling is not a reliable guide to its accuracy, especially outside your areas of genuine expertise.
The opposite rule — 'always analyse everything' — is not well supported either. Endless deliberation over small, reversible choices wastes energy, and in domains where you do have valid expertise, overriding good intuition with laborious analysis can make decisions worse, not better. More thinking is not uniformly better thinking.
Treating 'gut versus head' as a fixed personality trait does not help. The research frames it as situational, not dispositional: the same person should trust intuition in their domain of expertise and distrust it in a novel high-stakes call. Identifying as 'an intuitive person' or 'an analytical person' encourages applying one mode where the other fits better.
Real numbers in context
There is no single statistic that settles this, and any source claiming intuition is right 'X% of the time' should be treated with suspicion — accuracy depends entirely on the domain. The robust finding is conditional, not numerical: Kahneman and Klein concluded that valid intuitive expertise requires an environment regular enough to be predictable plus quick, adequate feedback to learn from. Both conditions present, intuition is trustworthy; either missing, it is not.
The practical takeaway is to classify the decision, not to score your gut. Familiar, fast-feedback, low-stakes, reversible — a quick intuitive call is usually fine. Novel, rare, high-stakes, irreversible, or probability-driven — the evidence favours deliberate analysis, ideally cross-checked against intuition. The mode that wins is the one that fits the situation, and the situation is usually identifiable before you decide.