Getting started isn't the problem. Surviving the first six months is.
I've watched people try to turn freelancing into a career more times than I can count. Some make it. Most don't, and it's rarely because they couldn't do the work. It's because they treated it like a hobby instead of a business from day one. There is a big difference. Here is how to actually approach it when you're coming from a full-time job with nothing but a laptop and a skill you're decent at.
How To Freelancing Tutorial
Pick a niche before you pick a platform
Most beginners skip this. They sign up on Upwork, Fiverr, or whatever the current hot board is, and immediately start bidding on everything. Data entry. Logo design. WordPress fixes. Translation. Voiceover work. It looks like a strategy. It isn't. It's a fast track to becoming the cheapest option in every category. The people who actually build stable freelance careers narrow down to something specific within the first month. "I write email sequences for B2B SaaS companies" is a position. "I write content" is a complaint waiting to happen. Specificity attracts the right clients and repels the ones who will waste your time. I learned this the hard way. My first year I took anything that paid. I ended up doing logo designs at $50 each when my actual rate should have been closer to $400 for that scope. A client once asked me to redesign their entire brand identity for what amounted to a coffee budget. I said yes because I was desperate. That project ate three days and paid less than minimum wage. Never again.
Set your floor rate and stick to it
You need a number. Not a vague idea of what you want to make per month. A hard hourly or project-based floor that you will not go below under any circumstances. Calculating this takes about ten minutes. Take your desired annual income. Divide by 1000 billable hours. That's your break-even number. Billable hours means the actual hours you spend working for paying clients, not the total hours in a year. Most freelancers bill between 800 and 1200 hours annually after accounting for admin,work, taxes, and the inevitable sick days. Use 1000 as a default if you aren't sure. Then add 30 percent on top for taxes, health insurance if you're in the US, software, and everything else that isn't glamorous but exists. That final number is your minimum. Any project below it is a bad deal unless there is a strategic reason to take it.
Get the Full Details

There is a strategic reason occasionally. Taking a project at a lower rate for a well-known company can get you a referral or a case study. But that has to be intentional. It can't just be you being scared to say no.
Build a portfolio before you need one
This is one of those counter-intuitive things. You think you need a client to get a portfolio. You actually need a portfolio to get a client. The solution is simple enough: create three to five spec pieces that demonstrate exactly the kind of work your ideal client would hire you for. If you're a web developer, build a complete landing page for a fake restaurant. If you're a copywriter, write a full email sequence for a fictional product. If you're a graphic designer, create a branding package for a pretend company. These don't need to be perfect. They need to look real and solve real problems. Put them on a simple site. Carrd or Framer works fine. Don't overthink the design. The work inside matters more than the wrapper. A bare-bones site with strong samples beats a beautifully designed site with nothing to show.
Where to find the first clients
Freelance marketplaces exist. They work for some people. They are brutal for most others because you're competing directly on price with millions of people worldwide. I don't recommend them as a primary strategy. Use them as a temporary bridge, not a destination. Better paths for the first few clients: Reach out to people you already know. Former colleagues. Friends of friends. Someone who runs a small business and might need help. These conversations are awkward at first. They get easier. I got my first three paying clients this way. One was a former coworker who needed a website update. The other two came from that same person after she saw the finished work.

Cold outreach to small businesses. Pick a type of business in your area or niche. Find the owner's email. Send a short message explaining what you do and asking if they need help with anything specific. Keep it under 100 words. Don't attach a resume. Don't link to a nine-page portfolio. Just state what you do and ask if they need anything. Contribute in public. Post your work on LinkedIn, Twitter, or relevant subreddits. Not in a braggy way. Just share what you're building or fixing. People notice. I once posted a before and after of a landing page I redesigned for practice. Got three inquiry messages within 48 hours. One converted into a $2000 project.
The unglamorous stuff that kills freelancers
It's not the work. It's everything around the work. Invoicing. Contracts. Follow-ups for payment. Negotiating scope changes. Dealing with clients who want revisions that weren't in the original agreement. Get a simple contract template before your first paid project. There are free ones from sources like Docracy or the Freelancers Union. Customize it for your situation. The most important clauses are scope, payment schedule, and revision limits. Without these, you will lose money on projects that slowly expand beyond what was agreed upon. Never start work without a deposit. 50 percent upfront is standard for projects over $500. For smaller jobs, get at least 25 percent. I once did a full branding project without a deposit because the client was a friend of a friend. They paid late. Then they asked for three extra rounds of revisions on top. It took six weeks instead of two. Lessons are expensive when you learn them the wrong way.
When freelancing stops working for you
There are honest limits to this path. Some people burn out within a year because the income fluctuates too much. Some niches are saturated to the point where you need years of reputation before you can charge decent rates. Some personalities thrive in structured environments and simply don't have the temperament for the constant self-management freelancing requires. If you find yourself struggling after six months of consistent effort, it might not be a skill problem. It might be a mismatch between how you work and how freelancing demands you work. In that case, a part-time arrangement or a role-based contract position through a platform like Toptal or Gun.io might suit you better than pure freelance chaos. Also consider that freelancing is a business model, not a personality trait. People who treat it like a business survive. People who treat it like a dream tend to run out of money by month eight. The line between those two approaches is usually just a spreadsheet and the discipline to follow it.

A note on pricing psychology
New freelancers often price too low because they think it makes them more attractive. It does the opposite. Clients who shop on price alone are the worst clients. They complain the most, pay the least, and never refer anyone. Pricing at or above market rate filters for people who value the work and are willing to communicate clearly. If you're unsure what the market rate is for your skill in your location, check job postings on LinkedIn and Indeed. Look at what companies are offering for similar roles. That gives you a baseline. Then adjust based on your experience level and the specific demands of your target clients. There is no shortcut around the early grind. But there is a shortcut around making the same mistake twice. Write things down. Track your hours. Keep a record of every conversation, every change request, and every payment received. When a client tries to claim you promised something you didn't, you'll have a paper trail instead of a argument.