Should you quit your job to start a business? Write the threshold first

Not yet. Almost certainly not yet, and the more useful version of this question isn't should I but what would have to be true before I did.

Write three numbers and a date. A monthly profit floor the business has to clear. The months of personal runway you will still be holding on the day you resign. And the date you sit down and check both.

That's an exit threshold. It's the same discipline as writing your kill criteria before you build: you decide while you're still capable of being fair, then you execute the decision instead of arguing with it at 11pm after a bad Tuesday.

The rest of this post is how to pick the three numbers, what the evidence says about leaping versus laddering, and the one thing this threshold can't tell you.

Everything ranking for this question is a feeling

We pulled the first page of Google for "should I quit my job to start a business" on July 24, 2026. The top result is a Reddit thread. Below it: career blogs offering signs, lessons, and steps.

Add "quiz" to that search, which is a real thing people type, and the page fills with personality tests. Are you unhappy at work. Are you a risk taker. Do you have what it takes. One of the ranking results is a LinkedIn post whose title abandons the genre halfway through: "Should I Quit My Job? Take This Quiz to--Just Kidding".

Nothing on either page gives you a number.

That isn't the writers' fault. The honest answer depends on your rent, your dependents, your savings, and what your business actually earned last month, and no article can know those. What an article can do is hand you the instrument and make you fill it in yourself.

The ladder beats the leap, and there's evidence

The romantic version of this decision is the leap. Burn the boats, full commitment, no plan B. It tells well at a dinner party. It also describes the riskier path.

Joseph Raffiee and Jie Feng looked at exactly this question in Should I Quit My Day Job?: A Hybrid Path to Entrepreneurship, published in the Academy of Management Journal (57(4), 936–963, 2014), using decades of US National Longitudinal Survey of Youth data. They compared hybrid entrants, people who started a business while keeping their wage job, against people who went straight from paid employment into full-time self-employment. Businesses started the hybrid way survived longer. The stretch spent running both at once also improved the odds of surviving once those founders did go full-time.

Read the direction rather than a precise effect size. It's observational data, so more cautious people plausibly sort themselves into the hybrid path in the first place. But the folklore says the leap signals conviction and conviction wins, and the data points the other way.

Keeping the job is not a failure of nerve. It's a longer runway, bought at the price of slower weeks.

And staying got cheaper. When shipping a product took six months of evenings, the day job really was the binding constraint. Now that building collapsed to a weekend, the thing your evenings can't manufacture is demand. And demand is public enough to read before you commit, which is worth doing while resigning is still hypothetical.

Write the threshold

Three numbers and a date. Fill this in tonight, before anything is at stake:

My exit threshold

  Profit  ≥ $______ / month, three consecutive months,
            earned while I am still employed

  Runway  ≥ ______ months of household costs, in cash,
            on the day I resign

  Check on: ____ / ____ / ______

If all three are not true on that date, I do not quit.
Next check date: ____ / ____ / ______

1. The profit floor. Use profit you actually keep, not revenue, and not a valuation. Set it as a fraction of your take-home pay rather than all of it, because you're not replacing your salary on day one, you're buying the right to try. Then the clause that does the real work: it has to clear that floor while you're still employed and part-time. If you have to quit for the numbers to work, you are betting that undivided attention creates demand. It normally doesn't. Attention creates output, and you already established that output is cheap.

   Three consecutive months, because one strong month is noise. Two is a coincidence you'll be tempted to spend.

2. The runway floor. Months of household costs, in cash, on resignation day, not counting the business's money and not counting credit. Pick the number of months from how long your market takes to answer a question, not from a round figure that sounds brave. If your sales cycle runs 60 days, three months of runway buys you roughly one and a half attempts at anything.

3. The date. An actual calendar date, chosen now. This is the part everyone skips, and it's the part that converts a wish into a decision. Without a date the threshold is never met and never failed, so it turns into something you'll get to.

Then the last line, which matters as much as the other three: what happens when the threshold isn't met. Not "keep going and see." You name the next check date, and you go back to work.

What each outcome actually means

The threshold gives you four results, not two, and three of them are informative.

Row three is the expensive one, because it's the row where quitting feels defensible. You have savings, you're miserable at work, the math on paper says you can survive a year. All true, and none of it evidence that anyone wants what you're building.

Write it while you still have nothing invested

The threshold only works if it predates the feelings it's meant to survive.

By the time you're seriously considering resigning, you've told people. You've described the business at three dinners and watched them get excited on your behalf. You have a name you like. Every one of those is a reason to revise the number downward, and each revision will feel like refinement rather than retreat.

The same trap sits under kill criteria written after you've built something, and it's why the fix in both cases is procedural rather than emotional. Decide early. Execute later. Don't negotiate in between.

Asking a chat model at that point won't rescue you either, for a structural reason we've covered: it validates you, not your idea. By the time you're typing the question, you've already told it how much you want the answer to be yes.

And if you hate your job, say so out loud and separately. It's a real problem and it deserves a real solution. It is not evidence about your business, and a threshold is the thing that keeps the two apart when a bad quarter at work makes them feel like one decision.

What the threshold can't tell you

It can only tell you whether you can afford to go. It cannot tell you whether the thing is worth going for.

Those are two decisions with two different failure modes, and running them in the wrong order is how founders end up with a fully funded year spent on an idea that was never going to work. Test the idea first. Then price your exit from it.

That first test is what MakeOrKillIt does. It runs five must-pass gates before it scores anything, then six weighted dimensions, and it returns a range with a confidence rather than a fake single number. One of those gates is resource availability: whether you have the money, time, and access to reach a first paying customer. That gate is as close to your personal life as the audit gets, and it stops there on purpose. It doesn't know your rent, your mortgage, or how many people depend on your salary. You do, and that's your half of the work.

So run them in order. Find out whether the idea deserves a threshold. Then write the threshold, with a date on it, tonight.

A resignation you can't justify in numbers is just a mood with a notice period.

Score an idea — free