Freelance Cash Flow: What to Do When You're Profitable but Have No Money

Published July 31, 2026

Disclosure: Some links below are affiliate links. If you sign up through them we may earn a commission at no extra cost to you. We only recommend tools we'd genuinely suggest to a friend. See our full disclosure.

You had a good few months. The work came in, you did it well, the invoices went out. Your records show a healthy profit and your tax pot is filling up. And then a Tuesday arrives where the software renewals and the accountant’s bill land in the same week, and there is not enough in the account to cover them.

Nothing went wrong. No client refused to pay. Every one of them paid, or will pay, exactly on the terms you agreed. You are not badly off — you are owed a great deal of money. You just do not have any.

This is the most common financial problem in freelancing and almost nothing written for freelancers describes it, because the entire money conversation is about how much you earn and how much you keep. Both of those are questions about amounts. This one is a question about when.

The quick version

• Profit and cash are different questions. Profit is a verdict on a finished period. Cash is a question about next Tuesday.
• The gap is longer than “30-day terms” implies. Count from the day you start the work, not the day you send the invoice, and add every step in between.
• A busy month widens the gap, because more work means more hours and costs incurred before any of it turns into money.
• Your monthly review will never catch it. It is backwards-looking by design; a cash gap is a fact about the future.
• The fix is a forward view, not more discipline — one page, money in and money out by date, and the number you care about is the low point, not the total.
• “Build three months of savings” is right and useless as a first move, because a buffer is built out of a surplus you do not yet have.
• The fast moves are administrative, not commercial: invoice immediately, ask the boring onboarding questions, move money to the front of the job.
• Solving a cash gap by taking work you would otherwise decline is the most expensive borrowing available, and it never appears as a cost anywhere.

Two numbers that are not the same number

Good freelance bookkeeping advice — including ours — tells you to log income and expenses as they happen, watch profit rather than revenue, and review monthly. That is correct, and it is the right foundation. It is also, structurally, a record of things that have already occurred.

Profit tells you whether a period you have finished living through was worth it. It puts the cost of a job and the payment for that job in the same reckoning, because for the purpose of judging the work, they belong together.

Your bank account does not experience them together. It experiences the costs in March and the payment in May.

That is the whole problem, and it is not a failure of discipline or of record-keeping. It is a difference in what the two numbers are for. A freelancer can be genuinely, sustainably profitable and still be unable to pay a bill in a specific week — not as a warning sign, but as ordinary arithmetic. The books are not wrong. They are answering a different question.

The gap is longer than you think

Most freelancers, asked how long they wait to be paid, will name their payment terms. Thirty days, say. But terms measure from the invoice date, and the invoice date is nowhere near the start of your exposure.

Run the real clock. Count from the first day you did work you had not been paid for:

• The work itself. A two-week project is two weeks of your time and any costs it needed, spent before an invoice exists at all.
• The drift between finishing and invoicing. Almost everyone has some. The job ends on a Thursday, you are straight into the next one, and the invoice goes out the following Tuesday.
• The agreed terms. Now the official clock starts.
• Their process. This is the step nobody accounts for. Your invoice may need approving by a specific person, or carrying a reference number, or arriving before a cut-off to make that month’s payment run. Larger organisations often pay on fixed days rather than on your due date, which means “due on the 19th” can genuinely mean “paid on the 24th” with nobody at fault.
• The transfer. Small, but it is not zero, and it lands after everything else.

Stack those and a client on thirty-day terms is routinely a two-month client measured from the day you started working. Do the sum with your own last completed job — the date you started, the date the money actually appeared — and you will have a more useful number than any rule of thumb, because it is the real behaviour of the real clients you have.

The reason this matters is that the gap is structural. It is not the result of anyone being slow. Shortening it is possible, and we will get to how, but you cannot eliminate it, and any plan that assumes money arrives when the work is done is planning for a business other than the one you have.

Why the good months are the dangerous ones

Here is the part that catches people, because it inverts the obvious expectation.

A quiet month is uncomfortable but visible. You can see nothing is coming; you behave accordingly.

A busy month feels like the opposite of a problem, and it widens the gap. More work means more hours spent before invoicing, more costs incurred up front, and — if you have taken on more than usual — possibly some of it subcontracted or bought in, which you pay for on someone else’s terms rather than your own. Everything you did in a big month becomes cash later. Everything it cost you was cash now.

So the squeeze routinely arrives one or two months after the best stretch of work you have had. That timing is exactly wrong for noticing it, because your recent memory says the business is going well, and it is. This is a different problem wearing the same clothes as success.

It is also worth separating from the related trap of work arriving unevenly. Having too much work is about the volume of jobs swinging around; this is about the money swinging around even when the volume does not. You can have a perfectly steady workload and still hit a cash gap, purely because of when the payments land relative to the bills.

Nothing in your normal routine will warn you

Think about what you actually look at.

Your bookkeeping shows the past. Your invoicing shows what you have sent. Your project tracker shows what is outstanding, usually as a chase-list — a set of things to follow up on, not a set of dated amounts arriving on a timeline. Your bank balance shows today, which is the one day you are never worried about.

Not one of those is a forward view. Every tool a freelancer normally has answers “what happened” or “what is open”. None of them answers “what will this account look like on the 14th of next month”.

That is why the gap is discovered rather than predicted. Almost everyone who has hit one found out on the morning of, from the bank app.

The forward view: one page, two columns

This is the entire fix, and it is smaller than it sounds. You are building a schedule, not a system.

Take the next twelve weeks. On one side list money out, by date:

• Fixed costs that leave whether or not you work — software, hosting, phone, anything with a recurring charge.
• Annual charges falling in that window. These are the ones that ambush people, because an annual renewal appears in a monthly review as a single odd-looking month rather than as a commitment. If you have never listed them, auditing your subscriptions is the fastest way to find them all at once.
• Tax payments, if any fall due in the window.
• Anything personal that has to come out of the business, if you do not draw a fixed amount.

On the other side list money in, by the date you realistically expect it, not the date you invoiced:

• Only work that is actually contracted. Not proposals, not likely renewals, not “they usually come back in spring”. A cash schedule that includes hopeful income is a wish list, and it will comfort you at precisely the moment you needed to be uncomfortable.
• Date each amount by your real clock from the section above, including the client’s payment run if you know it.

Then run a balance forward from what is genuinely in the account today. Add the money in on its date, subtract the money out on its date, and keep a running total.

You do not need software for this. A sheet of paper works for twelve weeks. A spreadsheet is better only because it re-totals when a date moves, which it will. If you already keep your income and expenses in a sheet, adding a third tab with these two dated columns is a ten-minute job — and if you would rather not build it, our Business & Freelancer Income & Expense Tracker already totals income, expenses, profit and tax set-aside in one place, which gives you the historical half; the dated forward column is the piece you add to it. Prefer paper? The free printable income & budget tracker covers the same ground by hand, no signup.

The number that matters is the low point

Once the running balance exists, ignore the total. The total is a profit-shaped number and you already have one of those.

Look for the lowest figure the running balance reaches, and the date it happens.

That is your actual constraint. It is the moment your business is at its thinnest, and it is almost never the month you would have guessed, because it is produced by the interaction of two schedules rather than by any single event.

Three things follow immediately from having that number:

1. If it is negative, you now know the date. Weeks in advance, with options — which is an entirely different situation from finding out on the day, when your only options are bad ones.
2. You know how much you are short by, which tells you the size of the move you need. Sometimes it is one invoice sent a week earlier. That is a very different problem from needing a month’s income.
3. You can test decisions against it. Can I afford this? is unanswerable in the abstract and trivial once you can see what the balance does on the date the money leaves.

Re-run it monthly, in the same slot where you already review your books. It takes minutes once the list exists, because most of the lines do not change.

Six moves, in order of speed

The standard advice is to build a buffer of several months’ costs. It is correct. It is also useless as a first move, because a buffer is made of surplus and the whole problem is that you do not currently have one. So here are the moves that work before you have savings, fastest first.

1. Send the invoice the moment the work is done. Every day of drift between finishing and invoicing is a day added to the wait, and it is the one delay entirely within your control. It is also completely free. If you do nothing else on this list, do this one — most freelancers can find a week here without changing anything about how they work. Our guide to invoicing a client covers what the invoice needs to contain so it does not come back.

2. Ask the boring administrative questions before you start. Not “what are your payment terms” — that is on the contract. Ask who receives the invoice, what reference or purchase-order number it needs to carry, and when payments actually go out. Three questions, one email, asked during onboarding when goodwill is at its highest and nobody has anything to be defensive about. An invoice that bounces back for a missing reference does not just lose you the days it sat there; it restarts the clock.

3. Move money to the front of the job. A deposit before work starts, or staged payments on a longer project, changes the shape of your cash flow more than anything else on this list, because it converts the worst version — all your costs first, all the money last — into something you can survive. Write it into the contract rather than raising it later; it is a normal professional term, not a favour. It has a second benefit that has nothing to do with cash: a client who will not pay a deposit has told you something useful before you have done any work, which is one of the signals worth reading early.

4. Shorten your terms on new work. Your current terms are probably not a decision. They are whatever was typed into a template once. You cannot easily change them on an existing client without making it an event, but every new client is a free opportunity to set a shorter number, and nobody negotiates payment terms at the start of a relationship they are keen to begin.

5. Move your own outgoings, not just theirs. You have been treating the money-out column as fixed. Some of it is not. Annual renewals can often be timed, and if three of them currently fall in the same month, that month is doing damage that has nothing to do with your clients. When something comes up for renewal, you get to choose which month it lives in from then on.

6. Split every payment on arrival. The habit of moving a tax percentage the moment you are paid is already the right instinct — extend it. Send a fixed, small percentage of every payment somewhere you do not spend from, on the day it lands. A buffer built from a percentage of income arrives whether or not you had a good month; a buffer built from “what’s left over” never gets built, because there is never anything left over.

And then, over a longer horizon: income that arrives on its own schedule rather than a client’s. A small recurring arrangement, a digital product, anything that pays in a trickle rather than a lump, does something no amount of invoicing discipline can — it puts money in the money-in column that is not waiting on a project to finish. This is also the single best reason to keep a small standing arrangement alive even when project work is plentiful, which is the same logic that protects you from depending on one client.

The most expensive way to close a cash gap

When the low point is a fortnight away and it is negative, there is one move that always seems available: take work. Any work. The enquiry you would normally decline, at the rate you would normally refuse, with the client whose first message already told you how the project would go.

Understand what that actually is. It is borrowing, at a rate you never calculate, from your own future capacity. You will pay it back in rework, in the hours a badly-defined job eats, in the enquiry you cannot take next month because you are still finishing this one, and in the rate you have now demonstrated you will accept to a client you will hear from again.

None of that shows up as a cost anywhere in your books. It is entirely invisible to the profit number and entirely real.

The reason to build the forward view is not the tidiness of it. It is that seeing the gap four weeks out instead of four days out is the difference between having options — send one invoice early, delay one renewal, ask one client about a deposit on the next stage — and having only this one.

Three things not to do

Do not chase people who are not late. There is a real difference between slow and late, and it is worth being precise about, because collapsing them costs you relationships for nothing. A client sitting inside agreed thirty-day terms on day twelve is not doing anything wrong; following up damages a relationship and gains you nothing, because the problem was the terms rather than their behaviour. Genuine lateness is a separate matter and deserves the polite escalation ladder — start it on the day after the due date, not before.

Do not make early-payment discounts a habit. In a genuine emergency, offering something off to be paid this week can be worth it. As a standing policy it is a very expensive loan, and it is worth doing the arithmetic once so you can see that: work out what the discount costs you, how many days earlier it brings the money, and what that would be as an annual rate. The quieter cost is that after a few rounds, the discounted figure is simply what you charge that client now.

Do not fund a gap with credit you cannot date. Bridging a specific shortfall with a card or an overdraft, when you can point at the exact payment that clears it and the date it arrives, is a decision. Doing it without being able to name that payment is not bridging a gap, it is starting a second one, and it will arrive with interest attached and no clearing date to plan around.

When it is not a cash-flow problem at all

Three cases where everything above is the wrong tool, and all are worth checking before you spend a month optimising your invoicing.

If the money is not coming because a client will not pay, that is non-payment, not timing. Different problem, different response — you need the escalation process and, eventually, the terms in your contract, not a forward-view spreadsheet.

If you work at full capacity, manage the gap properly, and still cannot build any buffer at all, then the timing is not what is wrong. You have a pricing problem wearing a cash-flow disguise. No amount of invoicing faster fixes a rate that does not clear your costs with room to spare, and the honest test is whether a fully-booked month leaves anything behind. If it does not, what you charge is the thing to fix, and every hour spent on payment logistics instead is an hour spent making the wrong number arrive sooner.

If a fixed amount already arrives from one client on the same day every month, your timing problem may already be solved — and something else may have replaced it. A predictable monthly payment for general availability is a relief after a year of lumpy income, and it is also one of the ways a freelance relationship quietly takes the shape of a job. Worth knowing which of the two you have bought.

The bottom line

Freelance money has two clocks. One measures whether the work was worth doing, and your books already read it accurately. The other measures whether the money arrives before the bills do, and almost nobody reads it at all, because no ordinary part of running a freelance business ever displays it.

The gap between those two clocks is not a sign of failure and it is not evidence that anyone treated you badly. It is the standard shape of the arrangement: you fund the work, and you get paid for it later. Once you can see it written down — twelve weeks, two columns, a running balance, one low point — it stops being an ambush and becomes a scheduling problem, which is a far more manageable kind of problem to have.

Build the page. Find your low point. Then work on the fast moves, in order, and let the buffer build itself out of a percentage rather than out of good intentions.

Next: how to invoice a client, freelancer bookkeeping made simple, and what to do when one client is most of your income. And if you sell products rather than hours, there is a second place money can stall that no forecast will show you — sitting with a platform that has suspended your account and is holding the balance.