There is a folder on your computer, or a notebook in a drawer, or a half-built thing in a corner of your apartment. You know exactly what it is. You haven't touched it in weeks, maybe months. You don't enjoy it anymore, and if you're honest, you're not sure you ever believed in it as much as you pretended to. And yet you can't throw it out, and you can't say the words I'm done. Every time you almost do, a small voice says: but you've already put so much into it.

That voice is not wisdom. It is one of the most thoroughly documented errors in human decision-making, and it may be quietly costing you your best hours.

What the sunk cost fallacy actually is

A sunk cost is anything you've already spent that you can't get back: money, time, effort, emotional energy. The fallacy is letting that spending steer a decision about the future, when the only thing that should matter is what happens next.

The classic demonstration comes from Hal Arkes and Catherine Blumer in 1985. They sold theater season tickets at randomly assigned prices: full price, a discount, or a bigger discount. Over the first half of the season, the people who had paid full price attended more plays than those who got the discounts. The seats and the shows were the same, but the pain of wasting a bigger payment pulled them through the door.

A related line of research by Barry Staw, beginning in the 1970s, studied what he called escalation of commitment. People who had made an earlier decision, and were then shown it was going badly, often invested more in it rather than less. The bad outcome didn't make them reconsider. It made them dig in.

It's easy to see this in companies and governments. It's harder to see in your own life, because in your own life it doesn't look like irrationality. It looks like loyalty, grit, and not being a quitter.

Why quitting hurts more than it should

Several well-studied mechanisms stack up here.

Loss aversion. Kahneman and Tversky's work on prospect theory showed that losses loom larger than equivalent gains. Abandoning a project turns your past effort into a visible loss. Continuing keeps the loss abstract. You aren't choosing between "finish" and "stop." You're choosing between "maybe it works out" and "admit those months were wasted."

Waste aversion. Arkes later argued that people carry a learned rule against wasting things. It's a good rule most of the time. It misfires when applied to time already gone, because that time is wasted whether you continue or not. Continuing only wastes more.

Self-justification. Leon Festinger's cognitive dissonance theory explains the rest. If you've spent six months on something, then deciding it's not worth it means admitting that the person who spent those six months made a mistake. That person was you. Staying the course protects your self-image as someone who makes sound decisions.

Identity. When a project becomes part of how you describe yourself (I'm writing a book, I'm learning to code, I'm training for this), quitting feels less like closing a task and more like deleting a piece of yourself. That's a much heavier cost than the decision deserves.

Notice what these have in common. None of them involve the project's actual future. They are all about the past, or about how you feel about yourself.

The costume the fallacy wears

The tricky part is that the sunk cost fallacy rarely announces itself. It shows up in sentences that sound responsible:

  • "I can't stop now, I'm so close." (Are you? Or have you been "so close" for a while?)
  • "I've already paid for the course."
  • "If I quit, it'll all have been for nothing."
  • "I just need to push through this rough patch."

Sometimes those statements are true. Persistence matters, and plenty of worthwhile things feel terrible in the middle. This is what makes the fallacy hard. The goal isn't to quit more. The goal is to make the decision using the right information, instead of letting the past cast a vote it shouldn't have.

The question that cuts through it

The most useful reframe is almost embarrassingly simple: if I were starting from today, with everything I know now, and none of the history, would I begin this?

Not "would I be willing to continue," which is contaminated by everything you've invested. Would you begin, given what's left to do, what it will cost, and what else you could do with that time?

If the answer is a clear yes, keep going, and you can stop worrying that it's a fallacy. If the answer is no, or a reluctant maybe, then the only thing holding you in place is the past.

There's one more distinction worth making. Sunk effort is not always lost. Skills, understanding, and clarity about what you actually want are things you keep. Quitting a project doesn't delete what it taught you. It simply stops adding to the pile.

Decide when you're calm, not when you're attached

Research on escalation suggests that people are worst at ending things while they're in the middle of them. So the strongest defense is to decide before you're emotionally entangled, in the same spirit as the commitment devices and pre-mortems you may already know.

This is sometimes called setting a kill criterion: a condition, written in advance, that tells you when to stop or change course. "If I haven't had a single paying client by the end of March, I'll rethink the plan." "If I've missed three of my four weekly writing sessions for a month, I'll review whether this still matters to me." The point is that the earlier, calmer version of you makes the call, and the later, defensive version only has to check whether the condition was met.

Your next moves

  • Pick the one project you're avoiding and run the clean-slate test today. Write one sentence: "Knowing what I know now, would I start this today, yes or no?" Don't edit your answer to be kind to your past self.
  • List what you'd keep, separately from what you've spent. On paper, make two columns: skills and knowledge I gained and time and money I can't recover. Only the first column belongs in the decision. Let the second stay where it is.
  • Write a kill criterion with a date. For anything ongoing, set a specific, observable condition and a review date on your calendar. "If X hasn't happened by Y, I stop or change the plan." Decide it now, while you're calm.
  • Do a 25-minute honest session before the verdict. If you can't tell whether you've lost interest or just hit a hard stretch, work on the project for one focused block with no multitasking. Afterward, notice whether the difficulty feels like stuck-but-interesting or flat and dutiful. Those feel different.
  • Tell one person the stop-or-continue decision you're facing. Not for permission, but so the choice is out loud. Quiet, private commitments are the easiest to drag on forever.

Where a tool can help

The part Tally can honestly claim is small: it makes the checkpoint cheap. Because focus sessions are bounded and your habits are stacked into routines you already follow, a short review of "is this still worth my hours?" can live as a regular step in your day instead of a dramatic reckoning you keep postponing. The decision to quit or continue is always yours. A structure that makes it visible, in small and regular doses, just keeps the past from deciding for you. If that appeals, you can look at Tally. And if it doesn't, the clean-slate question works just as well on a napkin.