Starting things is easy. Stopping them is where judgment goes to die.
A project that isn’t working rarely announces itself. It drifts. The numbers are a little short, but there’s a reason. The timeline slips, but the next month looks better. Every week you’ve put in becomes another reason to put in one more, because walking away would mean admitting those weeks were spent on something that didn’t work.
By the time the evidence is undeniable, you’ve spent far more than the thing was ever worth.
There is a simple defense, and it costs about ten minutes: decide your exit conditions before you begin.
Annie Duke calls these kill criteria in her book Quit: a state you can observe and a date by which you will check it. Here is how to write them for a small business.
Why we can’t quit well in the moment
When you’re inside a project, three forces push you to continue.
You’ve already invested, and abandoning it feels like losing that investment, even though it’s gone either way.
You’ve told people about it, and stopping feels like a public failure.
You can always imagine the version where it turns around, and that version is vivid while the slow decline is not.
None of these forces care whether continuing is a good idea. They operate regardless. Which is why the decision about when to stop shouldn’t be made by the person in the middle of the project. It should be made by the calmer person at the start.
How to set exit conditions
Before you commit to anything substantial, such as a new offer, a hire, a partnership or a side project, write down three things.
What success looks like, and by when. One or two concrete measures with a date. Not “growing,” but “twenty paying customers by the end of March.”
The point at which you’ll stop. A specific threshold, written as a sentence you can’t argue with later. “If we have fewer than eight paying customers by the end of March, I stop and move the time elsewhere.”
What you’ll do with the freed-up resources. This matters more than it looks. Quitting is much easier when you know exactly where the time and money will go next.
Then put the date in your calendar.
The middle zone
Most outcomes land between success and the stop line. Fewer customers than hoped, more than the minimum. This is where honest judgment matters most, and where it’s easiest to fool yourself.
Set a rule for this zone too. A good default: if you’re in the middle at the checkpoint, you get one extension, of a fixed length, with one specific change. Not “keep going and see.” Something like: “Six more weeks, with a different price, then the same test.”
One extension. Not two. The second extension is where most slow failures live.
Tell someone
Exit conditions are much stronger when someone else knows them. Not to hold you to account in a punishing way, but so there’s one person who’ll ask, gently, “It’s the end of March. Where did you land?”
That question, asked by someone outside your head, cuts through most of the stories you’d otherwise tell yourself.
This weekend
Look at what you’re currently working on. Pick the one thing you started without clear exit conditions. Write the three sentences now: success, the stop line, and where the resources go. Put the checkpoint in your calendar.
You might find it’s already past the line. That’s useful too.
What’s the one project you’d write a stop line for today? Hit reply and tell me. I read every one.
Also from Nolan Saye, in Flame Spiral: When a Project Deserves to Die, the same question for creative work: ending a project on purpose instead of letting it drift.
— Nolan


