Start from the job, not from the budget
Four numbers decide your budget. You already have three of them, and the fourth is the one most owners have never written down.
| Input | What it means | Where to get it |
|---|---|---|
| Job profit | What is left from an average job after materials, labour and vehicle | Your accounting software, averaged over a long enough run of jobs to be honest |
| Acquisition ceiling | The most you will pay to win one job | A decision, not a lookup — usually a share of job profit you are comfortable giving up |
| Jobs wanted | How many extra jobs a month you can actually service | Your capacity, not your ambition |
| Enquiry-to-job rate | The share of enquiries that become paid work | Count last quarter's enquiries and last quarter's jobs from those enquiries |
Multiply the ceiling by the jobs you want. That is what you are willing to spend to get them. You do not need to divide by anything else: the enquiry-to-job rate is already baked into what a job costs you, provided you measured the ceiling against booked work rather than against enquiries.
A worked example, with placeholder figures
These figures are invented for the arithmetic only. Substitute your own before you draw any conclusion from them. Say a job leaves you $600, you decide you will give up a third of that to win one, and you want ten more jobs a month. Your ceiling is $200 a job, so your monthly spend is $2,000. If one enquiry in four becomes a job, you need forty enquiries, which means you can pay up to $50 an enquiry. Now you have two numbers to manage against instead of a feeling.
The moment those two numbers exist, the rest of the account has a job description. Everything is measured against cost per booked job, and the measurement setup stops being a technical detail and becomes the thing the budget depends on.
Your market sets a floor, and you do not get to choose it
A click costs what your competitors are prepared to pay for it. In categories where one job is worth a great deal — legal help, emergency trades, anything with an urgent problem behind it — clicks are dear because several serious businesses want the same person. In quieter categories they are cheap. Neither fact is under your control.
This produces the most common failure in small budgets: the calculated number is real, but it is below the cost of being meaningfully present in the market the business has chosen to chase. Spread thinly across a whole city and every service on the van, the money buys a scattering of clicks and no conclusion.
When the honest budget is smaller than the cost of presence, narrow the market. Do not thin the money.
Narrowing is the correct response, and it is usually the unpopular one. Fewer services, a smaller area, or the times when your phone actually gets answered — any of those concentrate the same money into a market where it can register. Use Keyword Planner inside the account to see forecast costs for your own terms in your own area before you assume a national average applies to you.
If the numbers still do not work after narrowing, that is useful information rather than a failure. It means paid search is not where this business gets its next ten jobs, and the money belongs in a different part of the acquisition programme.
What Google actually does with the daily budget you set
The number you enter is an average, not a cap. Google Ads Help documents that spend on any individual day can run above your average daily budget when there is more demand than usual, and that the amount you can be charged across a monthly period is capped at your average daily budget multiplied by 30.4.
Two practical consequences follow. The first is that a single expensive day is not evidence of anything going wrong, and an owner who checks the account daily will be alarmed regularly for no reason. The second is that the monthly figure is the one that is genuinely controlled, so the monthly figure is the one to reconcile against your bank feed.
If you are registered for GST, the cost that belongs in your calculation is the amount net of the credit you can claim, not the gross charge on the card. That is a conversation for your accountant, but it moves the arithmetic enough to matter on a small budget.
A budget is not a spending decision. It is a price you are offering to pay for a booked job.
How much money does it take before the number means anything?
Budgets are usually judged far too early. If a month produces three enquiries, the difference between one job and two jobs from them is the difference between a triumph and a disaster, and it is entirely luck. You have not learned anything about the market. You have learned about a coin toss.
Before you commit, work out roughly how many enquiries your budget should produce. If that number is small, decide in advance how long you will run before you judge it, and hold to that. Changing the budget every week restarts the read and guarantees you never get one.
Three symptoms of a budget judged too early
- The spend went up or down more than once inside a fortnight
- The verdict changed after a single good or bad enquiry
- Nobody can say how many enquiries would count as enough to decide
Agreeing the review window before the money goes in is the cheapest discipline available, and it is the one most often skipped. Search campaigns are the fastest channel to read, and even they need more than a fortnight when volumes are modest.
When to increase, when to hold, and when to stop
Increase when two things are true at once: cost per booked job is comfortably under your ceiling, and you have capacity to service more work. Only one of those is not a reason. Raising spend to fill a diary you cannot service produces angry customers and bad reviews, which costs more than the campaign ever earned.
Hold when cost per booked job is sitting at the ceiling. That is the market telling you the price, and pushing more money in usually raises the price rather than the volume.
Stop when cost per booked job has been above the ceiling across a window long enough that it is not noise — and only after you have checked that the measurement did not change. A tracking change and a performance change look identical in a report. Rule out the cheap explanation first.
One more failure mode worth naming: the budget set from what is left over at the end of the month. It moves constantly, it can never be judged, and it quietly guarantees that the account is never given the run it needs to prove or disprove itself. Decide the number, write it down, and let it sit still long enough to answer the question.
If you want to see what the far end of this looks like when the numbers are published rather than described, our written-up engagements state the budget, the window and the conversion definition beside the result.
If the honest number is smaller than what it costs to be present, the answer is a narrower market, not a thinner budget.
Common questions
Is there a minimum budget for Google Ads in Australia?
Google sets no minimum. Your market sets a practical one. If your budget only covers a handful of clicks a day, you will not gather enough enquiries to tell a good result from a lucky one. The fix is a narrower market, not a thinner budget.
Should I spend a percentage of revenue on advertising?
Percentage-of-revenue rules are a planning convenience, not a decision. They tell you what you can afford and never whether the spend will pay. Work back from the value of a job and your conversion rate, then check the answer against what you can afford.
Will Google spend more than my daily budget?
On an individual day, yes. Google Ads Help documents that daily spend can exceed the average daily budget, and that what you can be charged across a month is capped at that average multiplied by 30.4. Judge the spend over a month rather than a day.
How long before I know whether the budget was right?
Long enough to collect a readable number of enquiries. If the spend produces two or three enquiries a month, no amount of patience makes that data reliable. Concentrate the budget on a smaller market, or accept that the read will be slow and decide the review window in advance.
What if the leads are cheap but bad?
Then cost per enquiry is the wrong measure and you should stop managing to it. Move the target to cost per booked job, which requires the outcome of each enquiry to be recorded and fed back. That is a lead quality problem, and it is fixable without touching the budget at all.