How to Start Freelancing in Pakistan: An Honest First 90 Days
Most freelancing guides sell you a timeline that does not exist. This one gives you the real 90 days — what you spend before you earn, what the fee stack takes off every dollar, and the month where most people quit.
Almost every guide to freelancing in Pakistan makes the same two promises: that you can start with nothing, and that the money starts within weeks. Both are close enough to true to be quotable and far enough from true to be useless as a plan.
This guide is built the other way round. It starts from what the first 90 days actually cost you, in rupees and in hours, and works forward to when income realistically appears. There are no earnings figures promised anywhere in this article, because nobody — including anyone who tells you otherwise — can predict what you specifically will earn. What can be stated precisely is the cost side, the fee side and the timeline. Those are the parts you can plan around.
First, the part nobody budgets for: the fee stack
The single most common surprise for new Pakistani freelancers is the gap between the number on the invoice and the number that lands in a local bank account. It is not one fee. It is a stack of them, applied in sequence.
Work an example through the whole chain. A client agrees to pay $100 for a piece of work:
| Stage | Typical deduction | Left |
|---|---|---|
| Invoice agreed with client | — | $100.00 |
| Platform commission (10% on the major marketplaces) | −$10.00 | $90.00 |
| Withdrawal to a payment provider | −$0 to $2.00 | ~$88.00–90.00 |
| Payoneer withdrawal fee to a Pakistani account (3%) | −$2.64 | ~$85.36 |
| Exchange-rate margin below the interbank rate (~1–2%) | −$0.85 to $1.70 | ~$83.66–84.51 |
So roughly 15–17% of a headline invoice does not reach you. Plan on keeping about 83–85% of what you agree with a client, and treat anything better as a good month.
Two of those lines moved recently and are worth watching. Payoneer raised its withdrawal fee for Pakistani users to 3% — a 50% increase on the previous rate — which alone costs about $2.64 on a $100 invoice. And the exchange-rate margin is the fee almost nobody counts, because it is never presented as a fee; it is simply a slightly worse rate than the one you would find by searching for the dollar rate that morning.
The practical consequence for pricing: if you need a specific rupee amount, divide, don't multiply. Work out the rupees you need, divide by 0.84, and that is the dollar figure to quote. Quoting the rupee equivalent directly leaves you about a sixth short, every single time.
What the first 90 days actually cost
"Start with zero investment" is the standard claim. It is not quite a lie — no platform charges you to open an account — but there are unavoidable costs, and pretending otherwise is how people end up stopping in week three because something broke and there was no money to fix it.
| Item | Realistic cost | Optional? |
|---|---|---|
| Reliable internet (fibre or a decent broadband package) | Rs 2,500–5,000/month | No |
| Backup connectivity (mobile data package) | Rs 500–1,500/month | No — power and line outages are the norm, not the exception |
| Power backup for outages (UPS, or shared solar) | Rs 15,000–60,000 one-off | Depends entirely on your area's load-shedding |
| A machine that can do the work | Already owned, or Rs 40,000+ used | No |
| CNIC (needed for payment account verification) | Already held | No |
| Portfolio samples — your own time | 20–40 unpaid hours | No |
| Paid courses | Rs 0 | Yes — skip these at the start |
The line that matters most is the second one. A missed client deadline because the electricity went and the UPS was flat is worth more damage than the cost of the UPS. Freelancing sells reliability before it sells skill, and in Pakistan reliability has an infrastructure cost that freelancers in other countries simply do not carry.
The line to ignore is the last one. Every rupee spent on a "freelancing course" in month one is a rupee spent before you know what you actually need to learn. The platforms' own help documentation is free, current, and written by the people who enforce the rules.
The 90-day timeline, week by week
This is the sequence that matches how the platforms actually work — profile review queues, buyer behaviour, payment clearing periods. It is not an aggressive timeline. It is a realistic one.
Days 1–14 — Pick one skill and prove it
Choose one service you can already deliver at a professional standard, or can reach within two weeks. Not three. Buyers search for specialists and the profiles listing eight unrelated services read as someone who does none of them well.
Then build three portfolio pieces. Real constraint, real brief, finished to the standard you would hand a paying client. If you have never had a client, invent the brief — take a real local business's existing material and produce your improved version, clearly labelled as a self-directed sample rather than as client work. Never present unpaid samples as paid engagements. It is checkable, and being caught costs you the account.
Days 15–21 — Accounts and verification
Open your marketplace profile and your payment account in the same week, because verification is the step that silently eats time. Getting a payment account approved, verified and linked to a Pakistani bank account takes days, not minutes, and you do not want to discover that after a client has already paid you. We walk through the whole process in the step-by-step Payoneer setup guide for Pakistan.
Your profile needs a real photo of your face, a specific title, and a description that states what you deliver rather than how passionate you are. Compare:
- "Hardworking and passionate designer, available 24/7, 100% client satisfaction guaranteed." — says nothing, matches ten thousand identical profiles.
- "I design product-page graphics for e-commerce sellers. Source files included, two revisions, 48-hour turnaround." — says exactly what a buyer gets.
Days 22–45 — Proposals, and the rejection you should expect
This is the stretch that decides who is still freelancing in month six. You will send a lot of proposals and hear nothing back on most of them. That is the normal experience of a profile with no reviews, not evidence that you chose wrong.
What measurably helps: applying only to jobs you can genuinely do, opening with the client's specific problem rather than your own biography, and keeping it to a few short paragraphs. What measurably does not help: applying to everything, long templated introductions, and the phrase "I am the best fit for this job" with no reasoning attached.
Set the expectation honestly. The first paid job is the hardest one in your entire freelancing life, because it is the only one you have to win with no track record at all. Every subsequent job is easier.
Days 46–75 — First jobs, priced for reviews
Price your first two or three jobs low enough to win and high enough that you are not resentful halfway through. You are buying reviews, and a review is the asset that makes every later proposal work. What you must not do is price at a level you intend to be permanent — raise your rate as soon as you have three good reviews, and expect to lose some price-sensitive clients when you do. That is the system working.
Over-deliver on the deadline rather than on the scope. Delivering early is remembered and costs you nothing; silently adding unpaid extra work teaches the client that your scope is negotiable after the fact.
Days 76–90 — First withdrawal, and the reality check
Money does not arrive when the client clicks approve. There is a platform clearing period — commonly around two weeks after order completion before funds become available — then the withdrawal itself, then the bank credit. Budget for the first rupees to reach your account roughly three weeks after your first job is approved, not the same day.
This is precisely why the fee stack matters and why you should not quit a paying job on the strength of one good freelancing month. Run the numbers into an actual budget before you make any decision that depends on the income — our monthly budgeting guide is built for exactly this, and for irregular income the emergency buffer it describes should be three months of fixed costs rather than one.
Choosing what to sell
The honest filter is the intersection of three things: something you can already do or learn in weeks, something with real demand, and something where you are not competing purely on being the cheapest person on the platform.
| Route | Time to first client-ready standard | Main risk |
|---|---|---|
| Graphic design (social, product pages) | Weeks, if you have any eye for it | Extremely crowded at the entry level |
| Writing and editing in English | Immediate if your English is genuinely strong | Buyers test hard; weak English is obvious in one paragraph |
| Video editing | 1–2 months | Needs a machine that can handle it |
| Web development | 3–6 months minimum | Long runway before the first rupee |
| Virtual assistance / data work | Immediate | Lowest rates, competes globally on price alone |
| Bookkeeping, translation, niche technical skills | Varies | Smaller market, but far less competition |
The bottom row is the underrated one. A narrow skill with modest demand and few qualified competitors beats a broad skill with enormous demand and fifty thousand competitors, particularly when you have no reviews.
Things that end accounts
Losing a platform account after building a review history is the worst outcome available to a freelancer, and it is nearly always self-inflicted:
- Taking the conversation off-platform to avoid commission. The 10% feels like theft until the client disappears without paying and you have no dispute process, no record, and no recourse.
- Multiple accounts. One person, one account. Device and payment fingerprints link them.
- Buying or trading reviews. Detected far more often than the people selling the service admit.
- Submitting AI output as original work when the brief forbids it. Increasingly checked, and a refund plus a bad review is the mild version.
- Location or identity misrepresentation. Verification is tied to your CNIC. There is no version of this that survives.
What to do this week
- Pick the one service you will lead with. Write it as a sentence a buyer would search.
- Block the hours for three portfolio pieces and put them in a calendar.
- Start the payment-account verification now — it is the longest-lead item.
- Work out your minimum acceptable rupee rate, divide by 0.84, and write the dollar figure somewhere you will see it before you send a proposal.
- Fix your backup power and connectivity before your first deadline, not after.
Freelancing is not the only route people try, and it is worth knowing how the alternatives compare on speed. Building a channel pays far later than client work does — at Pakistani audience rates it takes roughly 1.68 million monthly views to reach Rs 100,000, as the RPM arithmetic sets out.
Frequently asked questions
How much can I earn freelancing in Pakistan?
No honest answer exists at the individual level, and any specific figure you are quoted is either survey data about other people or marketing. What can be said is what determines it: your skill level, whether you are competing on price or capability, hours available, and how long you have been accumulating reviews. Treat published averages as context, never as a forecast for you.
How long before I earn anything?
Plan for the first payment to reach your bank account around three months after you start, allowing for profile setup, the proposal stretch, the job itself, the platform clearing period and the withdrawal. Some people are faster. Building the plan around the fast case is how people run out of money in month two.
Do I need to register a business or pay tax?
Tax treatment of freelance income, registration thresholds and any export-of-services concessions change frequently and depend on your total income. This article does not advise on it. Speak to a tax practitioner or check the current FBR position directly before assuming either that you owe nothing or that you owe the standard rate.
Is Fiverr or Upwork better for beginners in Pakistan?
They suit different starting points — one is closer to selling a defined product, the other to bidding for projects — and the fee and withdrawal costs differ once money reaches a Pakistani account. That comparison is its own article and is coming in this series.
Do I need a paid course to start?
No, and buying one in month one is usually money spent before you know what you are missing. Learn the skill from free material, start, and consider paying for training later against a specific gap you have actually hit.
My English is weak. Can I still freelance?
For client-facing written work, weak English is visible immediately and will cost you jobs — that is worth being direct about. For design, video editing and many technical services it matters far less, provided you can communicate clearly enough to understand a brief and report progress. Choose accordingly rather than pretending the gap is not there.
Sources and method. Payoneer's 3% withdrawal fee for Pakistani users is as reported in Pakistani business press coverage of the change; platform commission stated at the 10% standard applied by the major freelance marketplaces at time of writing. The fee-stack worked example is an original calculation for this article using those published rates and a 1–2% exchange-rate margin assumption; your actual figures will differ by route, account type and currency corridor. Startup-cost ranges are typical Pakistani market prices as at August 2026. Clearing periods are as documented by the platforms at time of writing. Fees and rules change often — verify current rates in your own account before pricing work. No earnings are promised or implied. This is general information, not financial, tax or legal advice.
