You've set your kill criteria. You know what signals would tell you it's time to stop. And yet, months later, you're still negotiating with yourself. The signals appeared. You explained them away. You're still going.
This happens because criteria without enforcement are just wishes. The psychologist Peter Gollwitzer spent decades studying why people fail to follow through on intentions. His research reveals something counterintuitive: knowing what you want to do and actually doing it are governed by different cognitive systems. You need a bridge between them.
That bridge is what we'll call states and dates: the two-part structure that converts vague criteria into automatic action triggers.
A state without a date is a wish. A date without a state is just a calendar reminder. Together, they create a decision rule you can actually follow.
The Implementation Intention Research
Gollwitzer's work on implementation intentions showed that people who specify when, where, and how they'll act are two to three times more likely to follow through than people who only specify what they want to achieve. The format is simple: "If X, then Y" or "When X happens, I will do Y."
This works because it pre-loads the decision. When the triggering situation arrives, you don't have to deliberate. The response is already queued. Your cognitive resources aren't depleted by the decision itself—the decision was made in advance, when you were thinking clearly.
For quitting decisions, this translates directly into states and dates:
- State = a measurable condition you either hit or miss (the benchmark)
- Date = the specific time you evaluate it (the deadline)
The formula: "If I am not in [state] by [date], I will [action]."
Why You Need Both
Consider what happens when you only have one:
State without date: "I'll quit if I'm not profitable." When? Next month? Next year? Next decade? Without a date, "someday" thinking takes over. You're always almost there. The goal post keeps moving.
Date without state: "I'll reassess in six months." Reassess against what? Without a clear state to evaluate, you'll look at the situation on that date and find reasons to continue. The assessment becomes a vibe check, not a decision.
The Pattern in Practice
A client had been working on a business idea for two years. When asked about his exit criteria, he said: "I'll quit when it's clearly not working."
The problem was visible immediately. "Clearly not working" isn't a state. It's an interpretation that will be influenced by exactly the biases he's trying to overcome—sunk costs, identity protection, loss aversion.
We rebuilt it: "If I don't have 10 paying customers generating $5,000 monthly revenue by June 30th, I'll stop and take a job."
That's a state (10 customers, $5K revenue) with a date (June 30th). When June 30th arrives, there's nothing to negotiate. Either the state is true or it isn't.
Common Failure Modes
1. States that aren't actually measurable
"If things don't improve" isn't measurable. "If I don't feel better about it" isn't measurable. These create infinite wiggle room. You need numbers, behaviours, or binary conditions that an outside observer could verify.
2. Dates set too far out
Setting a review date eighteen months away might feel "reasonable," but it's often a way of not setting one at all. Research on planning fallacy shows we consistently underestimate how long things take and overestimate future progress. Shorter cycles with clear escalation paths work better.
3. No consequence attached
States and dates need an "or else." Without specifying what happens if the state isn't reached by the date, you've just created a reminder to feel bad, not a decision rule.
"You're not choosing a future outcome. You're choosing a future decision rule."
The Three Forms
Gollwitzer's research, combined with practical application, suggests three standard forms that work for quitting decisions:
Form 1: Threshold crossing
"If I am not in [state] by [date], I will [quit/pivot/escalate]."
Example: "If I don't have a job offer by March 15th, I'll expand my search to different industries."
Form 2: Effort-limited
"If I haven't achieved [state] by the time I've invested [amount], I will stop."
Example: "If I haven't completed 20 exposure sessions by the end of this month, we redesign the protocol."
Form 3: Investigate-fast trigger
"If I observe [warning signal], I will take [specific action] within [short timeframe] to clarify."
Example: "If my anxiety hasn't reduced at all after four sessions, I'll ask my therapist directly whether we should try a different approach."
The Investigate-Fast Protocol
Not every signal requires immediate quitting. Some require investigation first. The investigate-fast protocol handles this:
When you notice a concerning signal, you have a preset action to clarify the situation—and a short timeframe to do it. This prevents both premature quitting and indefinite negotiation.
Investigate-Fast Template
Signal: [What I noticed]
Clarifying action: [What I'll do to get more information]
Timeframe: [Within how many days]
Decision rule: [If the answer is X, I continue for one more cycle. If the answer is Y, I stop.]
Clinical Applications
States and dates are particularly valuable in therapy contexts where people often stay in approaches that aren't working because "it takes time" or "I need to try harder."
The problem is real: therapeutic change does take time, and some approaches require patience. But that reality can become cover for never actually evaluating whether something is working.
States and dates create structure without rigidity:
- Anxiety treatment: "If my avoidance behaviours haven't reduced by 30% after eight weeks of consistent practice, we'll reconsider the protocol."
- Habit change: "If I'm not maintaining three sessions per week after the first month, we'll examine what's blocking implementation."
- Relationship work: "If there's no measurable change in conflict patterns after twelve sessions, we'll discuss whether couples therapy is the right intervention."
Important Distinction
States must track values-aligned outcomes, not momentary discomfort. "If I still feel anxious" is not a valid state—anxiety fluctuates and avoidance feels better short-term. "If my behavioural range hasn't expanded" is a valid state.
The States and Dates Ladder
For complex commitments, a single state-date pair isn't enough. You need a ladder: multiple checkpoints with escalating consequences.
Building Your Ladder
Step 1: Name what you're protecting
What matters more than this particular goal succeeding? (Health, time, relationships, other opportunities)
Step 2: Choose 2-3 states
What measurable conditions would tell you this is working? Be specific enough that someone else could verify them.
- State 1: _____
- State 2: _____
- State 3: _____
Step 3: Assign dates
When will you evaluate each state? Shorter cycles are usually better.
- Date for State 1: _____
- Date for State 2: _____
- Date for State 3: _____
Step 4: Specify consequences
What happens if each state isn't reached by its date?
- If State 1 not met ? [investigate / minor pivot]
- If State 2 not met ? [major redesign / prepare to quit]
- If State 3 not met ? [quit / full stop]
The 14-Day Signal Practice
Before building a full ladder, train your pattern recognition:
For the next 14 days, every time you catch yourself thinking "maybe soon," "it'll turn around," or "I just need more time," write down:
- What state would settle this question?
- What date would force clarity?
Don't change anything yet. Just notice how often you negotiate with vague criteria—and how easily you could convert those negotiations into decision rules.
Why This Matters for What Comes Next
States and dates are the enforcement layer for everything in this series. Kill criteria without states and dates drift into suggestions. Monkey-first testing without dates becomes indefinite preparation. The concepts only work when they're operationalised.
In the next piece, we'll examine why staying put feels so natural even when it's costing you—the endowment effect and status quo bias that make the default option quietly steal your life.
The Decision Series
Understanding When to Persist and When to Pivot
This content is for educational purposes and does not constitute psychological advice. If you're struggling with significant decisions affecting your mental health, consider speaking with a qualified professional.