You know the number. You have known it for a while. Maybe it is a product line that has missed forecast six quarters in a row. Maybe it is the second office you opened in a city that never produced. Maybe it is a key hire you are now the only person in the building still defending or a flagship client who costs more to serve than the relationship pays. You’ve run the math on a napkin. Then you put the napkin down and said something like: The strategy is sound, it just needs more time. That’s the sunk cost fallacy in its natural habitat—and the first thing to understand is that it does not feel like a mistake. It feels like leadership.
I coach high-performers for a living, and the hardest conversation I have with them is never about what to start. It is about what to stop. Here is what the research actually shows, where it comes up short, and the protocol I use to get a smart person out of a position they have been defending for too long.
The Season Ticket Study Still Explains Your P&L
In 1985 psychologists Hal Arkes and Catherine Blumer published “The Psychology of Sunk Cost,” one of the most cited findings in behavioral decision research. At the Ohio University theater box office, 54 season ticket buyers were randomly sold the same subscription at three prices: the full $15, a $2 discount, or a $7 discount. Then the researchers collected ticket stubs to see who showed up. Across the first five plays, full-price buyers used an average of 4.11 tickets. The discount groups used 3.32 and 3.29. Same plays, same access. The main difference was how much money had already left each person’s account, and it shaped attendance for the first half of the theater season.
Two findings from other experiments in that paper should make every owner and operator sit up. First, people who had already paid into a project rated the same project as more likely to succeed than people who had not. The spend may not just change what you do; it may change what you believe is probable. Second, the effect was not reduced by having taken economics courses. Knowing the fallacy by name provided no measurable protection.
Why Knowing About the Sunk Cost Fallacy Doesn’t Save You
That last finding is the one worth slowing down for because it means this is not an information problem. You cannot read your way out of it. What is running underneath is a rule, learned early and applied where it does not belong: do not waste what you paid for. Walking away feels like personally authorizing a subtraction, so the mind holds the position—and quietly inflates the forecast to justify holding.
In fairness to the skeptics, a 2025 paper in Brain Sciences found that the written scenarios researchers often use to measure sunk cost thinking have poor reliability as a scale. That critique deserves to be named. Notice what it does not touch: the theater study where real people spent their own money and then deciding, play by play, whether to go. The finding this article rests on is the one made of actual behavior.
The Tell Is That You Feel Calm
Here is what I watch for as a coach because the people I work with are not confused and they are not uninformed. They usually have better data on the position than anyone else alive. The tell is that somewhere in the past year they stopped evaluating the position and started representing it. They became its attorney. An evaluator gathers information to make a decision. An attorney gathers information to defend one that is already made.
The strange part is how calm it feels from the inside. There is no panic in it. There is a steadiness that looks exactly like conviction from the outside and feels like being the only adult in the room. So run the language test. If your private case for staying sounds like “the thesis is sound, the timing was early, give it two more quarters,” notice that nobody in that sentence is deciding anything. Someone is defending a spend.
Name the Payback You’re Protecting
The money is gone whether you stay or go. That is what makes it a sunk cost. So what exactly are you defending? Not the dollars. You are defending the payback—the belief that everything already committed will come back to you if you just hold on long enough. That payback has never existed anywhere except as a forecast, and the Arkes and Blumer data shows exactly what happens to a forecast made by someone who has already paid in: It inflates.
Try this in writing. Complete the sentence: “If I hold on, I will get ___ back by ___.” Now read it out loud to a colleague you respect. If you cannot say it without hedging, you have just met the real reason you are still in the position. You are not protecting an asset. You are protecting a number your own spending produced.
An Exit Protocol That Doesn’t Require Courage
You do not need to become braver. You need a process that does not rely on bravery.
First, ask the zero-based question: Knowing everything I know today, would I enter this position at today’s price? If the answer is no, you are not investing anymore. You are commemorating.
Second, switch sides for one hour. You have the best data on this position. Write one page arguing the exit case as if you had been hired to argue it. Most people have never once used their expertise on the other side of their own position.
Third, move the decision. Hand the call to a partner, a board member, or an adviser who didn’t write the original memo. The person with the history argues. A person without the history decides.
Fourth, set tripwires on everything new. Before your next commitment, write down what result by what date triggers an automatic review. You will never have less sunk cost than you have on day one. It is the same muscle that decides a plan is ready to launch—I wrote about that side of the problem in Why You Never Feel Ready to Launch Your Business. Verdicts avoided at the start become verdicts avoided at the end.
Now pick the napkin back up. Read it like an investor instead of an attorney.
Featured image from MayaLab/Shutterstock
.jpg)
