Salary, Skill, Way Out: If None Moved in a Year, That Is Your Signal

The other day I was editing the script that auto-publishes my WeChat posts. Halfway through, my eye caught a line in the auto-trading script sitting next to it.

That line was a stop-loss: if a given month's loss goes past a set percentage, the system halts itself. It doesn't wait for my nod, and it won't let me restart it by hand on the spot.

I stared at that line for a while, and something a little absurd hit me. I had written this "quit once it reaches this point" line for the machine. For my own job, all these years, I had never written one.

That is what this piece is about: an exit line. Why someone still doing frontline dev work should write one for himself.

First, what an exit line is. It is a trader's term, and the meaning is plain: before you get in, decide in advance the loss level that makes you walk, with no renegotiating with yourself. You set it not out of pessimism, but because your judgment gets a discount while you are in the thick of it, so you need a line fixed ahead of time to hit the brakes for you.

Why does a job need this line too? Because there is one illusion that is very easy to fall into: mistaking "this operation is growing" for "I am growing."

The budget on your project went up. The team went from three people to ten. Your boss walks into negotiations with more weight behind him than last year. All of that may be real, but what grew is the operation's ledger, not yours. Your own ledger is three other things: the salary you take home, the skill you can put on the table and deliver to outsiders directly, and what is left in your hands if you actually leave.

These two ledgers can come completely apart. The bigger the thing gets, the more fluent you become inside it, but what you are fluent in is this one set of processes, this batch of people, this client's temper. Move somewhere else and most of that fluency takes a haircut, because it grew around the operation.

So why, when the ledger clearly does not add up, do people still stay year after year?

Half of it is loss aversion. In the 1979 paper where Kahneman and Tversky laid out prospect theory, there is a finding that has been confirmed over and over since: the sting of losing something is a good deal stronger than the pleasure of gaining the same thing, roughly twice as strong. You have already put two or three years into this operation, and those years themselves become a "cannot let it go to waste" thing. Leaving feels like writing them off in one stroke, so even when it does not pay off now, you want to stick around and claw a bit of it back.

The other half is "just a little longer." Maybe there is a raise at year-end. Once this project ships next year it will be fine. After this crunch it will be my turn. Each of these sentences, on its own, is not false. The trouble is there is always a next one. One quarter after another, and by the time you actually come to, you may have run alongside this operation for several years, and the ledger is still the same ledger.

Sunk cost plus "just a little longer" adds up to this: no external signal will ever make you stop. You can only wait to snap out of it one day on your own. And people rarely snap out of it.

So setting an exit line is, at bottom, swapping "snapping out of it" for "doing the books on schedule."

Set yourself a cycle. Once a year works well; shorter and you get pulled around by momentary moods and swings. When the day comes, take an honest look at whether three things moved over the past year: salary, the work you can deliver to outsiders on your own, and your options after leaving this operation.

If even one of the three has not moved in a year, do not rush to quit. Just put "should I keep going" properly on the table and think it through once, instead of letting it auto-renew by default. If none of the three moved, that question will not take long.

I only added this line for myself this year. The stop-loss on the system side has been written for almost two years; my own side was always "let's see how it goes, it's fine." Writing it down has one very practical benefit: when the day comes I actually have to sit down and run the numbers, and what comes out is a concrete conclusion, not a vague "eh, good enough."

When was the last time you actually sat down and wrote out "the point at which I walk"? Or is this a line you have never written at all.