Most entrepreneurs know, intellectually, when a pivot is needed, the market signals are there, and the metrics are honest. The conversations with customers keep pointing in the same direction. And yet something resists. The decision that should feel obvious is not getting made. Understanding why that happens is more useful than being told to trust the data.

It is not irrationality. It is self-justification.

Barry Staw’s foundational research in 1976 established what he called escalation of commitment — the finding that personal responsibility for an initial decision creates a powerful psychological pressure to continue investing in it, even as evidence accumulates that the direction is wrong. The mechanism is self-justification, not stupidity. When you are the person who made the original call, abandoning that call is experienced as an admission of error. The brain resists this not because it cannot see the evidence but because updating on it carries a psychological cost that accumulating more evidence does not.

The sunk cost effect is the financial expression of the same mechanism. Time, money, team, narrative — all of it feels like it will have been wasted if the direction changes. Arkes and Blumer’s research identified at least five psychological factors feeding this: loss aversion, waste aversion, effort justification, the framing effect, and the gambler’s fallacy. Following through on the original decision can be framed as success. Cutting losses gets framed as failure — even when cutting losses is the rational choice.

What I find worth noting here is that a failed replication of Staw’s original paradigm exists, and the sunk cost literature has genuine methodological complexity. The self-justification mechanism is well-evidenced across many studies. Some specific experimental designs are contested. Present it as a real and documented pattern, not a law of human nature.

The practical consequence for entrepreneurs is specific. A founder who has raised money, hired a team, and spent eighteen months building in one direction is not evaluating pivot evidence from a neutral position. Every piece of evidence for the pivot also carries the implication that the previous eighteen months involved a mistake. The evaluation of the evidence and the evaluation of the self have become entangled.

The pivot threatens more than the strategy

Morgan Grimes’ 2018 study in the Academy of Management Journal is the most important research I found on this topic — the only major empirical work examining directly how founders respond to pivot-relevant feedback. What he found is that the response involves two simultaneous processes: idea work and identity work. Founders do not just evaluate whether the business direction should change. They evaluate what changing it means about who they are.

Entrepreneurial identity — the sense of self built around being a person who is building this particular thing — is not incidental to the pivot decision. It is central to it. When the core idea needs fundamental change, the feedback threatens the narrative through which the founder has constructed their sense of purpose and competence. The business and the self have fused. That fusion makes the pivot feel existential in a way the evidence alone does not explain.

The more publicly the founder has committed to the original direction, the higher the psychological stakes. A founder who has pitched at conferences, been featured in press, and built a team around a specific vision has an audience for the pivot. Changing direction is not just a private strategic update. It is a public narrative disruption. The identity cost of pivoting in front of people who know the original story is meaningfully higher than the identity cost of pivoting privately — which is why so many pivots get delayed until the evidence is so overwhelming that the social cost of not changing finally exceeds the identity cost of changing.

The founder who keeps finding reasons the data is not conclusive enough is often not evaluating the data. They are managing an identity threat.

Why it feels wrong even when you know it is right

The third mechanism is the most uncomfortable because it persists even after the decision to pivot has been made. Changing direction requires holding two things simultaneously: that the original decision was the best available given what was known at the time, and that it now needs to be substantially abandoned. Both are true. The tension between them does not resolve through logic.

Festinger’s cognitive dissonance theory, published in 1957 and one of the foundational frameworks in social psychology, describes this precisely. When behaviour and belief conflict, the brain works to restore consistency — usually by adjusting the belief rather than acknowledging the behaviour was wrong. In the pivot context, this produces the specific experience of the decision feeling wrong even when the rational case for it is clear. The wrongness is not evidence that the pivot is a mistake. It is evidence that it violates narrative consistency, which produces cognitive dissonance regardless of the underlying logic.

What makes this particularly difficult is that the dissonance is most powerful when it is unexamined. The founder who holds the tension privately, without articulating it, experiences it as a persistent sense of wrongness that can look like intuition against the pivot. Articulating it — writing out the narrative explicitly, saying out loud “we learned X, which means we need to do Y” — reduces its power. The dissonance does not disappear, but it becomes something to examine rather than something to resist.

This is harder than it needs to be because of how failure gets framed

Entrepreneurial culture has made reasonable intellectual progress on the idea that pivoting is not failure. The lean startup framework, Eric Ries’ build-measure-learn methodology, the entire vocabulary of iteration and learning — all of it is designed to create conditions where directional change feels like calibration rather than collapse.

The psychological reality is that this reframe, however accurate, does not fully neutralise the mechanisms the research describes. Self-justification, identity threat, and cognitive dissonance are not addressed by telling someone that pivoting is actually smart. They are addressed by changing the structure of the decision itself.

The most evidence-based intervention I came across is what researchers call the counterfactual framing — presenting the pivot decision as a new decision rather than a reversal of the old one. The question is not “should I abandon what I decided?” It is “if I were a new CEO with no prior investment in this direction, what would I decide?” The two questions are logically equivalent. Psychologically, they are completely different. The first activates self-justification. The second removes the personal responsibility dimension that triggers it.

Netflix’s explicit internal policy of not spending any time protecting its DVD business when streaming became viable is a clean example of this in practice. The decision was framed prospectively — where should resources go given current evidence — rather than retrospectively — how do we justify changing what we said we were doing.

What the research actually points towards

Three things are worth taking from this.

First, the resistance you feel toward a pivot that the evidence supports is not a reliable signal about the pivot. It is a reliable signal about the psychological cost of the pivot. Those are different things, and confusing them is how founders stay in wrong directions for longer than the evidence warrants.

Second, the identity cost of pivoting is real and worth managing rather than ignoring. Framing the pivot as learning rather than failure is not spin — it is a legitimate psychological tool for reducing the narrative disruption that identity threat produces. Telling the story of the pivot clearly, to yourself and to others, reduces cognitive dissonance rather than amplifying it.

Third, the sunk cost trap is weakest when the decision is evaluated prospectively. The years already invested are gone regardless of what you decide next. The question that produces the clearest thinking is not what those years require of the future but what the current evidence recommends for it.

A book worth reading alongside this

Mistakes Were Made (But Not by Me) by Carol Tavris and Elliot Aronson is the most readable treatment of the self-justification and cognitive dissonance research that underpins this article. Their pyramid of choice model — showing how each commitment makes the next commitment easier and the first reversal harder — maps precisely onto what the escalation mechanism produces in the pivot context. For any entrepreneur who has found themselves defending a direction they privately know is wrong, it is the most direct and honest conversation with why that happens.

This article discusses psychological patterns documented in research on decision-making and entrepreneurial behaviour. It is not designed to identify, diagnose, or assess any psychological condition, and it is not a substitute for professional support. If the difficulty of making a significant business decision is significantly affecting your wellbeing or functioning, speaking with a psychologist can provide support that goes further than understanding the mechanism.

This article is for educational and informational purposes only. It is not a substitute for professional psychological advice, diagnosis, or treatment. If you are experiencing significant psychological distress, please consult a qualified mental health professional.

Sources: Staw, B.M. (1976), Organizational Behavior and Human Performance, 16(1). Arkes & Blumer (1985), Organizational Behavior and Human Decision Processes, 35(1). Grimes, M.G. (2018), Academy of Management Journal, 61(5). Festinger, L. (1957), A Theory of Cognitive Dissonance. Mantere et al. (2013), Academy of Management Journal, 56(1). Hafenbrack, Kinias & Barsade (2014), Psychological Science.