A
azizfromgroundwork
Guest
Most freelancers think about payment terms as a question of trust. Will this client pay? Do I seem difficult if I ask for a deposit?
There is a more useful way to look at it. Every hour you work before you are paid for it is money you have lent to your client, interest-free, with no security. The amount you are owed at any moment is your exposure: the value of work you have delivered minus the cash you have received. It rises as you work, falls when a payment lands, and it is exactly what you lose if the client disappears tomorrow.
Payment terms decide how big that number gets and how long it stays there. The difference between a good structure and a bad one is not a few percent. On the same project, it is four times as much money at risk.
Take one project and run it three ways.
A $12,000 build, ten weeks long, with the work spread evenly: $1,200 of work delivered each week. Exposure at the end of any week is:
A negative number means the client has paid ahead of the work. A positive number is your loan to them.
The most common default. You invoice $12,000 when the project ends at week 10, and the client pays 30 days later, around the end of week 14.
Peak exposure: $12,000, held for four weeks. For fourteen weeks, you have been financing the client's project, and for the last four the entire fee is outstanding. If they go quiet in week 12, you have lost three months of work, not one invoice.
The usual "fair" split. $6,000 arrives before you start; the other $6,000 is invoiced at week 10, net 30.
Peak exposure: $6,000. Half of Structure A, which is why the deposit feels like the fix. But notice the shape: for the first five weeks the client is carrying the risk, and for the last nine it is all on you. The deposit protects the start of the project, and the start is the part that rarely goes wrong.
Same $12,000, split into four payments of $3,000: a deposit before work starts, then one payment at the end of week 4, one at the end of week 7, and the last at handover in week 10. Milestone invoices are due in 7 days, the final one in 14.
Peak exposure: $3,000, and the project is fully paid two weeks earlier. Same client, same work, same total fee. A quarter of the risk of Structure A.
Put the three side by side, and a pattern falls out:
In Structure A, the only payment is $12,000, so that is what you can lose. In Structure C, no single payment is bigger than $3,000, so no moment in the project has more than about $3,000 at stake.
You can design for this directly. Decide the most you are willing to lose if a client vanishes, then size the payments so none of them is bigger than that. If your answer is $3,000 and the project is $12,000, you need at least four payments. The negotiation stops being about trust and becomes arithmetic.
Exposure has two dimensions: how much, and for how long. Freelancers negotiate hard on the first and accept whatever they get on the second.
Suppose the Structure A client pays 30 days late, in week 18 instead of week 14. Your $12,000 of exposure now lasts eight weeks after handover instead of four. If you would otherwise cover your own bills on a credit card at, say, 24% a year, carrying $12,000 for an extra four weeks costs about $220:
That is roughly the cost of knocking 2% off your fee, except nobody asked you to agree to it.
Now, run the same 30-day delay on Structure C. The late payment is the final $3,000, so the cost is about $55. Milestones do not just cap the risk; they shrink the damage of the most common thing that goes wrong.
Splitting the invoice is the easy part. Three details decide whether the structure holds when a project goes sideways.
Tie each payment to something the client can see. "Phase 2 complete" invites an argument about what complete means. "Staging site live with checkout working" does not. A milestone the client can verify by clicking a link is a milestone they cannot dispute.
Do not start the next phase until the last payment lands. This is the clause that turns a schedule into protection. Without it, a client who is two weeks late on milestone 2 is also two weeks into milestone 3, and your exposure is quietly back to Structure A levels. Put it in the contract in one sentence so pausing work is following the agreement, not making a threat.
Shorten the terms on everything but the last invoice. Net 30 on a milestone mid-project means you keep working while you wait, which adds a full phase of unpaid work to your exposure. Seven days on milestones is normal and rarely questioned.
Before you send your next proposal, take five minutes:
If the answer is no, the fix is not a better client. It is a different payment structure, and it costs nothing to propose.
I write Groundwork, a publication for people running a business of one. If you want the contract side of this- how to write milestone triggers and what to check before you sign- I put it together in Payment Milestones: A Schedule Example and Pre-Signature Checklist.
AI disclosure: this article was written with the help of AI.
There is a more useful way to look at it. Every hour you work before you are paid for it is money you have lent to your client, interest-free, with no security. The amount you are owed at any moment is your exposure: the value of work you have delivered minus the cash you have received. It rises as you work, falls when a payment lands, and it is exactly what you lose if the client disappears tomorrow.
Payment terms decide how big that number gets and how long it stays there. The difference between a good structure and a bad one is not a few percent. On the same project, it is four times as much money at risk.
The Setup
Take one project and run it three ways.
A $12,000 build, ten weeks long, with the work spread evenly: $1,200 of work delivered each week. Exposure at the end of any week is:
Code:
exposure = work delivered so far - cash received so farA negative number means the client has paid ahead of the work. A positive number is your loan to them.
Structure A: Paid on Completion, Net 30
The most common default. You invoice $12,000 when the project ends at week 10, and the client pays 30 days later, around the end of week 14.
| End of week | Work delivered | Cash received | Exposure |
|---|---|---|---|
| 2 | $2,400 | $0 | $2,400 |
| 4 | $4,800 | $0 | $4,800 |
| 6 | $7,200 | $0 | $7,200 |
| 8 | $9,600 | $0 | $9,600 |
| 10 | $12,000 | $0 | $12,000 |
| 13 | $12,000 | $0 | $12,000 |
| 14 | $12,000 | $12,000 | $0 |
Peak exposure: $12,000, held for four weeks. For fourteen weeks, you have been financing the client's project, and for the last four the entire fee is outstanding. If they go quiet in week 12, you have lost three months of work, not one invoice.
Structure B: 50% Upfront, 50% on Completion
The usual "fair" split. $6,000 arrives before you start; the other $6,000 is invoiced at week 10, net 30.
| End of week | Work delivered | Cash received | Exposure |
|---|---|---|---|
| 2 | $2,400 | $6,000 | -$3,600 |
| 5 | $6,000 | $6,000 | $0 |
| 8 | $9,600 | $6,000 | $3,600 |
| 10 | $12,000 | $6,000 | $6,000 |
| 13 | $12,000 | $6,000 | $6,000 |
| 14 | $12,000 | $12,000 | $0 |
Peak exposure: $6,000. Half of Structure A, which is why the deposit feels like the fix. But notice the shape: for the first five weeks the client is carrying the risk, and for the last nine it is all on you. The deposit protects the start of the project, and the start is the part that rarely goes wrong.
Structure C: Four Milestones
Same $12,000, split into four payments of $3,000: a deposit before work starts, then one payment at the end of week 4, one at the end of week 7, and the last at handover in week 10. Milestone invoices are due in 7 days, the final one in 14.
| End of week | Work delivered | Cash received | Exposure |
|---|---|---|---|
| 1 | $1,200 | $3,000 | -$1,800 |
| 3 | $3,600 | $3,000 | $600 |
| 4 | $4,800 | $3,000 | $1,800 |
| 5 | $6,000 | $6,000 | $0 |
| 7 | $8,400 | $6,000 | $2,400 |
| 8 | $9,600 | $9,000 | $600 |
| 10 | $12,000 | $9,000 | $3,000 |
| 11 | $12,000 | $9,000 | $3,000 |
| 12 | $12,000 | $12,000 | $0 |
Peak exposure: $3,000, and the project is fully paid two weeks earlier. Same client, same work, same total fee. A quarter of the risk of Structure A.
The Rule Underneath
Put the three side by side, and a pattern falls out:
Code:
peak exposure ≈ the largest payment you are waiting forIn Structure A, the only payment is $12,000, so that is what you can lose. In Structure C, no single payment is bigger than $3,000, so no moment in the project has more than about $3,000 at stake.
You can design for this directly. Decide the most you are willing to lose if a client vanishes, then size the payments so none of them is bigger than that. If your answer is $3,000 and the project is $12,000, you need at least four payments. The negotiation stops being about trust and becomes arithmetic.
Late Payment Is Worse Than a Low Rate
Exposure has two dimensions: how much, and for how long. Freelancers negotiate hard on the first and accept whatever they get on the second.
Suppose the Structure A client pays 30 days late, in week 18 instead of week 14. Your $12,000 of exposure now lasts eight weeks after handover instead of four. If you would otherwise cover your own bills on a credit card at, say, 24% a year, carrying $12,000 for an extra four weeks costs about $220:
Code:
$12,000 × 24% × 4/52 ≈ $222That is roughly the cost of knocking 2% off your fee, except nobody asked you to agree to it.
Now, run the same 30-day delay on Structure C. The late payment is the final $3,000, so the cost is about $55. Milestones do not just cap the risk; they shrink the damage of the most common thing that goes wrong.
What Makes a Milestone Actually Work
Splitting the invoice is the easy part. Three details decide whether the structure holds when a project goes sideways.
Tie each payment to something the client can see. "Phase 2 complete" invites an argument about what complete means. "Staging site live with checkout working" does not. A milestone the client can verify by clicking a link is a milestone they cannot dispute.
Do not start the next phase until the last payment lands. This is the clause that turns a schedule into protection. Without it, a client who is two weeks late on milestone 2 is also two weeks into milestone 3, and your exposure is quietly back to Structure A levels. Put it in the contract in one sentence so pausing work is following the agreement, not making a threat.
Shorten the terms on everything but the last invoice. Net 30 on a milestone mid-project means you keep working while you wait, which adds a full phase of unpaid work to your exposure. Seven days on milestones is normal and rarely questioned.
Run Your Own Numbers
Before you send your next proposal, take five minutes:
- Write down the project fee and how many weeks it will take.
- Divide the fee by the weeks to get your weekly work value.
- Sketch when each payment would actually arrive, terms included.
- Find the week where work delivered minus cash received is highest.
- Ask yourself whether you could lose that number and still pay your rent.
If the answer is no, the fix is not a better client. It is a different payment structure, and it costs nothing to propose.
I write Groundwork, a publication for people running a business of one. If you want the contract side of this- how to write milestone triggers and what to check before you sign- I put it together in Payment Milestones: A Schedule Example and Pre-Signature Checklist.
AI disclosure: this article was written with the help of AI.