don't quit your 9-to-5 for your side project until it passes this test

the most dangerous month to quit your job is the month your side project has its best month yet.

that's the month the dream feels real. one good invoice, one viral post, one client who pays like the salary you're tired of - and the math in your head quietly rounds one good month up to every month from now on. it isn't. a good month is a data point, not an income. the people who quit on a single green month are the ones you see back on linkedin nine months later, "open to work", quietly relieved nobody asks how the thing went.

this is for the version of you that's burnt out, working hard for someone else, and pretty sure the way out is a project of your own. it is. but the exit and the quitting are two different events, and doing them on the same day is how the exit fails.

tl;drdon't quit when your side project has a good month. quit when it passes the replacement test - when its boring, repeatable, after-tax income covers your real monthly cost of living, for long enough that you'd bet a year of runway on it. below is the test, the four numbers it runs on, and the trap that fakes a pass.

the replacement test

you have replaced your income when all four of these are true at the same time:

1. the floor covers the nut. your side project's worst recent month - not the best, not the average - clears your real monthly cost of living after tax.
2. it's boring to earn. the income comes from something you can repeat on a normal tuesday when you're tired, not from a fluke you can't explain or reproduce.
3. it has held. the floor has cleared the nut for long enough that it's a pattern, not a spike. a good quarter beats a good month; a good two beats a good one.
4. you have runway anyway. you could survive a year if all of the above turned out to be wrong, because months four through nine of going full-time are usually worse than the side-project months, not better.

a "yes" needs all four. most people fail on one and quit on the other three because the other three feel like enough when you're desperate to leave. the floor is the one that saves you, and the floor is the one the good month hides.

the four numbers it runs on

the test is only as good as the numbers you feed it. get these four right and the rest is arithmetic.

• the nut - your real, all-in monthly cost of living. rent, food, insurance, debt minimums, the annual bills divided by twelve, the irregular ones you pretend don't exist. not your aspirational lean-budget number; the number your bank statement actually shows over the last six months. most people lowball this by twenty to thirty percent, which is exactly the gap that sinks them.
• the floor - the worst month your side project has produced in its recent run, after you set aside tax. not the average. the average is a story you tell yourself; the floor is the month you'll actually have to live through.
• the spread - the distance between your best month and your floor. a big spread means the income is lumpy and you need more runway and a longer hold before you trust it. a tight spread means it's predictable, which is worth more than a higher-but-wilder number.
• the runway - months of the nut you have saved in cash. this is the one you build before you need it, on purpose, while you still have the salary doing the saving for you.

write them down. four numbers, one page. if you can't fill them in from real data, you don't have a replacement yet - you have a hope with good lighting.

the trap: quitting on your best month

here's the move that gets people, drawn out plainly.

your side project averages a third of your salary. most months it's quiet. then one month three things land at once and it clears your whole salary. the relief is physical. you've been exhausted for two years and here, finally, is proof. you hand in your notice on the wave of that month.

then the wave passes, because waves do. the next month is back to the average - a third - except now there's no salary underneath it. you burn runway to cover the gap, runway you maybe didn't fully build because the good month made you feel rich. stress goes up, not down, and stress is the one thing that makes you worse at the creative, uncertain work that going full-time demands. the project that was growing nicely on the side starts shrinking under the weight of needing it to pay rent this month.

the good month didn't lie about the ceiling. it lied about the floor. the test exists to make you stare at the floor while the best month is busy showing you the ceiling.

what to do instead, while you still have the job

the job isn't the enemy here. it's the cheapest startup capital you will ever get - it pays you to keep yourself alive while you build the thing that replaces it. use it on purpose:

• raise the floor, not the ceiling. spend your side hours making the boring income more repeatable, not chasing another viral spike. a dull system that pays every month beats a brilliant one that pays sometimes.
• build the runway with the salary. every month you're still employed, the paycheck can stockpile months of the nut. that runway is what lets you quit on a yes instead of a flinch.
• practise the full-time work part-time. the hard part of self-employment isn't the work, it's the lack of structure. learn to ship without a boss while you still have one to fall back on.
• lower the nut where it's painless. every recurring cost you cut shrinks the number you have to replace. it's the same as a raise, and it's under your control today.
• use the leverage you have. a small ai-assisted stack lets one tired person deliver like a team, which is how you raise the floor without adding hours you don't have - how to make money with ai, and the realistic timelines in make money online in 2026.

none of this is the exciting part. that's the point. the exciting part is the trap.

when it's actually time

it's time when the test bores you. when the floor has cleared the nut for a few quarters, the income is something you can explain and repeat, the spread is tight enough that a bad month doesn't end you, and you've got a year of runway sitting there making the whole decision feel almost anticlimactic.

that's the feeling you're aiming for. not the high of a record month - the flatness of a number that's been true long enough to stop being exciting. quit on boring. boring is what replaced income feels like, and replaced income is the only thing worth quitting for.

build the thing on the side until it passes. then the day you leave is just paperwork, because the bridge already reaches the other side.

faq

when should i quit my job for my side project?

when its income passes the replacement test: the worst recent month clears your real monthly cost of living after tax, the income is boring and repeatable rather than a one-off spike, it's held for long enough to be a pattern, and you have around a year of expenses in cash. a single great month is not a pass - it's the most common reason people quit too early.

how much should my side project earn before i go full-time?

measure the floor, not the average. your worst recent month, after tax, should cover your real all-in cost of living - and it should do that consistently, not once. aim to leave with roughly a year of runway saved, because the first several months full-time are often leaner than the side-project months were.

why is a good month a bad reason to quit?

because a good month tells you the ceiling, not the floor, and you have to live on the floor. spikes don't repeat on demand. if you quit on a record month, the next ordinary month arrives with no salary underneath it, stress rises, and the pressure makes the work worse. wait until the boring, repeatable income - not the peak - covers your costs.

more: legit ways to make money online and how to make money with ai. more in the notes.