A client offers to pay you triple your hourly rate as a fixed employee, in exchange for invoicing as a freelancer. You do the math on a mental napkin: if you work 30 billable hours a week, in two months you'd already earn as much as an entire quarter as an employee. It sounds obvious. You quit on Monday.
The problem is that math is wrong, or at least incomplete. It doesn't include the months without a project, the health insurance you now pay yourself, the time you'll spend selling instead of coding, or what happens if that one big client decides to cut the contract in a three-line email. Before you stop being an employee, it's worth doing the full math, not the one that gets you most excited.
They're not the same thing: freelancing, consulting, and building your own product
"Starting your own thing" lumps together three paths with completely different risks and returns, and it's worth separating them before deciding anything.
Freelancing or contracting is selling your time, like as an employee, but to several clients instead of one. It's the lowest-risk path of the three: you still charge by the hour or by the project, with rates that for a senior LATAM dev working with US clients usually run from $35 to $90 an hour depending on stack, English level, and track record. There's a ceiling too: your income is still tied to available hours, so scaling means working more or raising your rate, not both indefinitely.
Consulting is a step further: you're not selling hours, you're selling results, and eventually you build a small team so you're not the bottleneck. The income potential is higher, but now you have to sell, invoice, chase late payments, and manage people, skills no coding bootcamp ever taught you.
Your own product (a SaaS, a developer tool, an app) is the most romanticized path, and statistically the one that works the least. Most indie products never take off, not because the code is bad, but because building is the easy part and getting your first hundred paying users is the hard one. If your plan is "I'll launch it and it'll go viral," it's worth rethinking before you quit your fixed salary.
The math almost nobody does: your real safety net
Before calculating how much you could earn, calculate how much you need to survive six months with zero income. That includes rent or mortgage, health insurance (which your company probably covered while you were a full-time employee), retirement contributions if your country requires them for independent workers, and a margin for months with zero invoicing, which will happen no matter what, especially in the first year.
A rough guideline: if you need $1,500 a month to live comfortably, your cushion before leaving your fixed job should be around $9,000, not the $3,000 that covers one bad month. That lets you absorb two or three months without invoicing without making desperate decisions, which are the ones that turn out worst.
Exchange rate risk also factors in if you invoice in pesos, reais, or any local currency while your big expenses (travel, equipment, some service billed in dollars) aren't. Getting paid in dollars as a freelancer for US clients is, in fact, one of the real reasons many LATAM devs prefer that path over starting a local consultancy.
What you really gain, and what it costs
The autonomy is real: you choose who you work with, when to take a day off without asking permission, and which projects to accept. Nobody's going to make you use a technology you don't believe in just because someone three levels above you decided it.
What gets talked about the least is the cost of operational solitude. As an employee you have a QA team, a support team, someone in people ops who handles your payroll questions, and a manager who absorbs part of the business's uncertainty. As a freelancer or consultant, all of that is you: you sell, you invoice, you do client support, and you code in whatever time is left over. A lot of devs underestimate how much of their week goes into tasks that aren't coding.
An alternative that rarely gets mentioned
Not everything that pushes someone to stop being an employee is the desire to own a business. A lot of the time it's the lack of control over your own work, a salary that doesn't reflect your real seniority, or the feeling that you've already hit the local growth ceiling. If that's your case, it's worth evaluating a middle path first: a remote job for a US company, with dollar compensation and more autonomy than a local position, before taking on the full risk of invoicing on your own.
It's not the right answer for everyone. If what you want is to answer to nobody, no remote job is going to give you that entirely. But if what you want is better pay, better treatment, and more say over technical decisions without taking on the full risk of selling and invoicing, it's a real option that a lot of devs skip straight past on their way to the fantasy of "quitting everything and starting a business."
How to test it without betting everything at once
Before quitting, take on a small freelance project alongside your current job, even if it's nights or weekends, for two or three months. That gives you real data on three things: whether you can land clients without relying on your current network, how much time selling and invoicing actually eats up, and whether you actually like the work itself, beyond the fantasy of freedom.
If those trial weeks go well and your cushion is in place, the decision stops being a leap of faith and becomes a calculated bet. If they go badly, better to find out while you still have your fixed salary than after you've already quit.
Leaving employment behind can be the best decision of your career, or a hard year that sends you back to job hunting with less savings than before. The difference between the two is almost never courage. It's the money you saved before jumping, and how honest you were with yourself about how long selling, not coding, is going to take.
