Exit Criteria: Why Your Future Self Needs a Pre-Written Permission Slip to Quit

You launched the side project fourteen months ago. The original plan was simple: if it hadn’t found paying customers within six months, you’d shut it down and redirect the energy. At month six, you had two users and no revenue — but you’d just redesigned the product and felt the new version was stronger. You gave it another three months. At month nine, one user had left, but you’d received encouraging feedback from a potential partner. You extended to twelve months. At month fourteen, the project has consumed hundreds of hours, several thousand pounds, and a meaningful portion of your identity. It still has no revenue. But stopping now would mean all of that was wasted.

At no point did you make a deliberate decision to continue past your original criteria. The criteria simply drifted — each extension felt reasonable, each “one more quarter” justified by some new signal. The goalposts moved so gradually that you never noticed they’d left the field entirely.

The research

Joel Brockner, in his 1992 review in the Academy of Management Review, identified goalpost drift as one of the core mechanisms of escalation of commitment. Decision-makers facing a failing course of action don’t typically make a single, conscious decision to continue. Instead, they make a series of small decisions, each of which feels marginal: a little more time, a little more money, a slight redefinition of success. Each increment is too small to trigger a fundamental reassessment. The cumulative effect is massive but invisible — a classic boiling-frog dynamic in which the conditions for quitting are never met because the conditions keep being redefined.

Barry Staw and Jerry Ross, in their 1987 review in Research in Organizational Behavior, identified the psychological drivers of this drift. Self-justification: continuing validates the original decision, while quitting implies the original decision was wrong. Completion proximity: the perception that success is “just around the corner” — a perception that renews itself at every stage. Identity investment: as time and effort accumulate, the project becomes intertwined with the decision-maker’s self-concept, making abandonment feel like a personal failure rather than a strategic decision.

Peter Gollwitzer, at New York University, published a landmark paper in American Psychologist in 1999 on implementation intentions — specific, pre-formed plans in the format “when situation X arises, I will do Y.” His research demonstrated that implementation intentions dramatically increase follow-through on goals and commitments. The mechanism is that the pre-formed intention creates a mental link between a trigger and an action, bypassing the deliberative process that would otherwise be required (and that would be vulnerable to rationalisation). Pre-set exit criteria function as a specific form of implementation intention: “when condition X is met, I will stop.”

Todd Rogers, Katherine Milkman, and Kevin Volpp, writing in JAMA in 2014, reviewed the broader literature on commitment devices — mechanisms by which people pre-commit to actions their future selves might resist. The finding was consistent: pre-commitments that are specific, time-bound, and linked to observable conditions outperform vague intentions by a wide margin. The power of the commitment device is that it removes the decision from the future moment — when rationalisation is strongest — and relocates it to the present — when judgement is clearest.

The mechanism

Richard Thaler and Cass Sunstein, in Nudge (2008), described the broader principle at work: people are systematically different decision-makers in “cold” states (calm, rational, uncommitted) and “hot” states (emotional, invested, committed). In a cold state, you can clearly see that a project with no revenue after fourteen months should be stopped. In a hot state — when you’ve invested your time, money, and identity — the same conclusion is almost impossible to reach. Exit criteria set in a cold state and enforced in a hot state transfer the authority to the version of you who can see clearly.

The specificity of exit criteria is essential because vague criteria are vulnerable to reinterpretation. “I’ll stop if it’s not working” gives your future self infinite latitude to define “working.” Each small positive signal — a compliment, a near-miss, a hint of progress — can be used to argue that it is, in fact, “working.” A specific criterion — “I’ll stop entirely if monthly recurring revenue is below £2,000 by September 30” — eliminates this latitude. The number is either met or it isn’t. The date either arrives or it doesn’t. The negotiation is removed.

Exit criteria differ from tripwires in a critical way. A tripwire says “reassess when this condition is met.” It opens a conversation. Exit criteria say “stop when this condition is met.” They end one. Both are valuable, but they serve different functions. The tripwire protects against drift. The exit criteria protect against the specific failure mode where reassessment consistently concludes “let’s give it one more quarter” — where the reassessment itself has been captured by escalation dynamics.

Your future self will be more invested, more attached, and more entangled than you are now. Exit criteria aren’t a constraint on that person. They’re a gift — a pre-written stop signal, from the only version of you who can grant it clearly.

The practical implications

Set exit criteria at the moment of highest clarity — before you start. The time to define what constitutes failure is when you have no skin in the game, no sunk costs to protect, and no identity wrapped around the project. Once you’ve started, every subsequent definition of failure will be biased by the desire to avoid triggering it. The present-tense version of you, uncommitted and clear-eyed, is the only reliable author of exit criteria.

Make the criteria specific enough that your future self can’t argue with them. “If things aren’t going well” is not an exit criterion — it’s an invitation to rationalise. “If monthly revenue is below £5,000 by Q2 end” is an exit criterion that either fires or it doesn’t. The specificity is the mechanism. Every dollar of ambiguity you leave in the criteria is a dollar your future self will use to justify continuation.

Separate the exit decision from the exit emotion. When the criteria are met, the emotion will say “one more try.” The criteria say “stop.” The exercise is in honouring the criteria despite the emotion — trusting that the version of you who set them was thinking more clearly than the version who is now trying to override them. This doesn’t mean the criteria are always right. But it means that overriding them should require a conscious, documented decision with new evidence — not a vague feeling that things might improve.

The bigger picture

The professional world celebrates persistence. The stories we tell about success — in business, athletics, creative endeavour — are stories of people who didn’t quit. What we don’t tell are the stories of people who persisted at the wrong thing for too long, draining resources and years that could have been redirected. For every narrative of grit rewarded, there are dozens of silent narratives of grit misapplied — and the difference between the two is often not the quality of the effort but the quality of the exit decision.

Exit criteria are an act of intellectual honesty that most professional cultures don’t reward. Saying “I’ll quit if X happens” feels defeatist. In practice, it’s the opposite — it’s the acknowledgement that you’re entering a situation with uncertainty, that you might be wrong, and that you’ve thought clearly enough about failure to specify what it looks like in advance.

The people who make the best long-term decisions aren’t those who never quit. They’re those who quit the right things at the right time — and the right time is almost always earlier than it feels. Exit criteria are the mechanism that makes “the right time” visible, even when everything inside you is arguing for one more try.