The reason freelancers take on too much is not greed or bad judgement. It is that the question gets asked the wrong way around. When an enquiry lands, most people check whether the new job fits in the calendar. The calendar is mostly empty, because meetings are the only thing that goes in it. The work you already owe is sitting in five other places.

Answering properly takes about fifteen minutes and gives you a number instead of a feeling. It also gives you the third answer, which is the one that keeps the client: not yes or no, but “yes, starting the 14th”.

Why does the calendar lie?

Because a calendar holds appointments, and freelance work is not appointments. A week showing two calls looks like a week with thirty free hours. In reality it already contains the second round of revisions on one project, a launch you promised to be around for, four invoices that need chasing, and the quarter’s paperwork.

None of those are in the calendar, so none of them are in the decision. That is the whole mechanism.

Step 1: Count what you already owe, in deliverables

Write down every outstanding deliverable across every client that lands inside the period you are being asked about. Deliverables, not clients, and not projects. “Acme website” is not a unit of work. “Two remaining page builds plus one revision round” is.

Include the things that are not yours yet but will be: a revision round you know is coming, a launch day where you have promised to be reachable, the second half of a job currently stalled on the client.

If this list does not already exist in one place, that is the real problem and it is worth fixing before the capacity question — we go through how in how to manage multiple clients as a freelancer.

Step 2: Turn it into hours using your own history

Do not estimate from scratch. Find the last two or three jobs of the same type and use what they actually took, including the revisions and the emails. Your estimate of a new page template is optimistic. Your record of the last four is not.

If you have never tracked this, start now with one number per job: hours from brief to sign-off, including admin. Three jobs is enough to be more accurate than your instinct. Ten is enough to quote from.

Step 3: Subtract the load that is never billable

This is the step that gets skipped, and it is worth more than the other four combined. In any month you also spend time on:

  • Quoting and proposals, including the ones that do not convert.
  • Invoicing, chasing, and reconciling.
  • Email and messages that are not attached to a deliverable.
  • Your own bookkeeping, and in Australia a BAS if you are registered for GST.
  • Whatever broke: a hosting issue, a file that will not export, a plugin update that took the afternoon.

Write down what that came to last month. Most people have never measured it once, and most people are surprised, because it is the part of the week that never feels like work and always takes the time.

The Australian version of this load

Two things move it predictably here. If you are registered for GST you have a BAS cycle, which is a fixed admin cost per period rather than a variable one. And if you lodge your own tax return as a sole trader, the self-lodgement deadline is 31 October, so October carries an admin spike that has nothing to do with client work. The current lodgement rules are on the ATO’s income tax return pages, and a registered tax agent can extend that date, which is one of the more valuable things an agent does for a sole trader.

Step 4: Apply a buffer you measured, not one you read

Everyone knows to leave a buffer and nearly everyone picks the number by feel. Measure it instead: over your last few projects, how much longer did they take than you said they would? If jobs consistently run 30% over, that is your multiplier until the evidence changes.

This number is personal and it is stable. It also does not mean you are bad at estimating. It means estimates are made in ideal conditions and delivered in real ones, where clients go quiet for nine days and come back wanting the header changed.

Step 5: Compare, then use the third answer

Now you have committed hours plus overhead plus buffer, against the hours that exist. If the new job fits, take it. If it does not, the useful move is almost never “no”.

“I’d like to do this. I can start on the 14th and have it to you by the 28th. If you need it sooner I can recommend someone, or we could do the first stage now and the rest from the 14th.”

That reply keeps you in the running, tells the truth, and gives the client two ways to say yes. A surprising share of urgent deadlines are urgent because nobody asked. The ones that are genuinely fixed will tell you, and then you have learned something real about the client rather than guessing.

What you must not do is take it and hope. The cost of a late delivery is not the apology, it is that it lands on the client who was already waiting, and you pay for the new work twice.

Capacity tells you whether there is room. Deciding what to do with the room, in a week where several jobs are already moving, is worked through in how to prioritise multiple client projects.

What about the clients you already have?

Capacity is not only about new enquiries. The same calculation answers the harder question: which existing client is quietly consuming a disproportionate share of your month?

Track two numbers per client over a quarter: hours worked and revenue. The ratio tells you things a gut feeling will not. The client who feels difficult is often fine. The client who feels easy but sends eleven small requests a month is usually the one costing you the next job. Counting the unbilled extras is covered in how to stop scope creep before it starts.

Does another client push you over the GST threshold?

This is an Australian question worth asking before you say yes rather than after. Once your GST turnover reaches $75,000 a year, or you expect it to, the ATO requires you to register within 21 days, and GST is 10%. The rules are on the ATO’s registering for GST page.

The point is not that crossing the threshold is bad. It is that it changes two things at once: your quotes need to handle GST consistently from that day, and your admin load gains a BAS cycle. Both belong in the capacity calculation, not in a surprise in three months. How GST changes the document you send is covered in how to write a quote for freelance work in Australia.

Frequently asked questions

How many hours a week can a freelancer realistically bill?

Fewer than the hours you work, always, and the honest answer is that you have to measure your own. Anyone quoting a universal figure is describing their own business. Track your billable hours for one month against total hours worked, and you will have a number you can plan with.

Should I take work I do not have capacity for and subcontract it?

Only if you have worked with the person before and you price for managing them. Subcontracting does not remove the work, it converts delivery hours into briefing, reviewing and fixing hours. On an unfamiliar collaborator that can be more expensive than doing it yourself.

What do I do when a client goes quiet and then comes back?

Treat it as a new scheduling request rather than a resumption. “Happy to pick this up — the next window I have is the 18th.” A project that stalled for three weeks on their side has no claim on the slot it used to hold, and saying so once, calmly, prevents it happening every time.

Is it worth keeping a waiting list?

Yes, if you actually contact the people on it. A short note saying “I have capacity from the 14th, is that still useful?” costs nothing and converts far better than a cold pitch, because they already wanted you once.

How do I know if I am underpriced rather than over capacity?

If you are fully booked and still short of money, it is a pricing problem, not a capacity problem, and taking another client will not fix it. Working out the rate that makes the maths work is covered in how much to charge as a freelancer in Australia.