Starting OutFreelanceYour first paid gig will most likely come from someone who already knows your name. A SelfEmployed.com guide on landing clients with no portfolio makes the same point: without ratings or history, you compete on price and visibility, and both favor people who have been around longer. Direct outreach to people you know usually produces better-fit clients faster.
Timelines are soft. Guides promise a first client in 2 to 6 weeks or 30 to 60 days, but none of them track a real cohort. Read those as the optimistic end.
Pick your skill before you pick your route. Generic blog writing is the weak spot right now. A study published in Management Science found freelance writing postings fell about 30% within eight months of ChatGPT’s launch, with software and design postings down roughly 21% and 17%. Later data is mixed: one analysis of 5 million Upwork listings found video editing up 39% and web design up 10%. Upwork’s February 2026 In-Demand Skills report says demand for skills tied to applying AI grew 109% year over year. Which skill could you sell by Friday? Keep that answer handy.
Route 1: pitch one narrow offer to people you know
This is the fastest route and the least glamorous. Former coworkers, classmates, neighbors, and your own LinkedIn contacts will take a chance on you before a stranger will, because they already trust you not to vanish.
The mechanism is simple. A vague message (”I do freelance stuff now”) gets sympathy. A specific one (”I will rewrite your five worst product pages for $250”) gets a yes or a no. You want the yes or no.
Write one sentence: who you help, what you deliver, what it costs.
Message 10 people individually. No group posts.
Ask who they know who needs this, not only whether they do.
Price by project. Guides put beginner hourly rates around $20 to $40, but a flat project fee keeps you from being paid less as you get faster.
Ask for a short testimonial the day you deliver.
The honest part: most of your friends are not your market. Expect a lot of polite silence. The referrals matter more than the direct yeses.
Route 2: build two proof pieces for real businesses
No portfolio does not mean no proof. It means you have to make some, and you do not need much. One guide says two finished pieces published today beat one polished piece that never ships, and that matches how clients behave. They want one or two signals that you can do the work.
Local businesses are the most overlooked source here. A dentist, a gym, or a bakery with a weak website or no email newsletter is a real brief with a real owner who can say this helped.
Pick two businesses with a visible, fixable problem.
Do the work for one of them free or at a token price, and cap it. One free job is an investment. Five is a hobby.
Get written permission to show the work and a two-sentence testimonial.
Publish both pieces on a single page you can link in a message.
Free work has a failure mode: the client who treats you as free forever. Set the scope in writing before you start. 💡
Route 3: send cold emails, but fewer than you think
Cold email works, just not at the rates people brag about. The numbers depend heavily on whose dataset you use. Across more than 850 million emails, Smartlead’s benchmark puts the typical sender at a 0.74% reply rate, or one reply per 135 contacts. Instantly’s 2026 report lands at a 3.43% average. Both are B2B sales data, not freelance data, and nobody publishes a clean freelance benchmark. Anyone quoting you 15% or 20% reply rates for freelancers is selling a template pack.
The same data points to what helps:
Under 80 words. Short first emails reply better in both datasets.
Small, hand-checked lists. Instantly data cited by SideStackers shows campaigns under 50 recipients averaging 5.8% against 2.1% for lists over 1,000.
A custom domain. Hunter’s data, summarized by Carly’s cold email statistics roundup, shows free webmail addresses getting 2.5% replies against 5.2% from a custom domain.
Follow-ups. About 42% of replies come after the first email, per Instantly.
Do the math before you start. Fifty emails at 0.74% to 3.43% is somewhere between zero and two replies. Interested replies are rarer still, around one in 157 contacts by one 2026 stat roundup. So you send 150 or more, personalized, to land one real conversation. Could you hand-write 50 emails this week? If not, pick another route. Worth knowing: the BizWhat Membership covers this topic in a dedicated ebook, which is a lot more useful than a single article can be.
Route 4: list one packaged gig on Fiverr
Fiverr flips the problem. You do not pitch anyone; you wait for buyers to find a listing. That sounds easy and mostly is not.
The cut is 20% of every order, tips included, at every seller level, per Fiverr’s terms as tracked by ConductAtlas. A $100 gig pays you $80. Buyers pay their own service fee of about 5.5% on top. New sellers also wait 14 days for earnings to clear, and get seven active gig slots.
The bigger issue is visibility. One 2026 review says zero reviews means zero search placement in most categories. Another site claims most new sellers earn $100 to $500 a month in their first 60 to 90 days, but that is one source, so treat it as a soft guess rather than a benchmark.
Sell one narrow, clearly bounded package, not “writing services.”
Build the 20% into your price from day one.
Use your Route 1 and Route 2 proof to get the first two reviews.
Send your own traffic to the gig. Waiting for search to find you is slow.
This route suits people with a defined, repeatable deliverable: a logo, a video edit, a set of product descriptions. It suits consultants badly.
Route 5: spend a few dollars on targeted Upwork proposals
Upwork charges to apply, which makes it the most expensive way to fail and a cheap way to test. Connects cost $0.15 each, sold in bundles starting at 10 for $1.50, according to several 2026 fee guides. Freelancer Plus is $19.99 a month with 100 Connects. Fee guides put the service fee at around 10% on most contracts.
The complaint on forums is not the price per Connect. It is that the number needed per job has climbed, so new freelancers burn through their starter batch fast. New accounts often get a small free allotment, and a one-time bonus of 50 Connects is reported for first purchases or Plus signups.
Set a budget and stick to it. Twenty dollars buys about 133 Connects, which is a handful of proposals, not a campaign.
Apply only to jobs posted in the last day or two.
Write the first two sentences about their problem, not your background.
Do not undercut everyone. A review of Upwork behavior notes that new freelancers consistently underprice, and the race to the bottom is crowded with people who already lost it.
Quote the project, not your hourly rate.
You may spend that $20 and win nothing. That is the real failure rate of this route, and it is why it goes last. If Routes 1 and 2 are done first, your proposals have something to point to.
Here is the one thing to do today: write your single-sentence offer, then message ten people you know before you go to bed. Who is the first name on your list?


