How Much Should You Spend on Google and Meta Ads? Work Backwards From One Customer
"How much should I spend?" is the first thing most owners ask about ads, and the usual answers are no help. "Start small and see." "Whatever you can afford." "10% of revenue." None of them know anything about your business.
There's a better way. Skip the budget for now and start with one customer. What are they worth to you? Once you know that, the budget mostly works itself out.
Step 1: Find what one job is worth
Take a Hamilton landscaper. An average job brings in $3,000. After materials, wages and fuel, about $1,000 is left over. That $1,000 is the number that matters, not the $3,000. It's the most he could ever pay to win that customer and still come out ahead.
Now he needs two more numbers, and he can get both from his own records:
Out of every 10 enquiries, how many turn into a paying job? Say 3.
Out of every 100 people who click his ad, how many send an enquiry? A ballpark figure for trades is 5 to 8, though it depends a lot on how good your website page is. Say 6.
Step 2: Do the sums
Here's the cost per click. One NZ pricing guide from this year puts trades at roughly $3.50 to $8.00 a click, and professional services like law much higher. Auckland runs about 25% dearer than elsewhere. Treat these as rough, since your own account will tell you the real figure within a few weeks. Say $5.
100 clicks cost $500.
Those clicks bring 6 enquiries, so each enquiry costs about $83.
3 in 10 enquiries become jobs, so each job costs about $277.
He pays $277 to win a job that leaves $1,000 behind. That's a good deal, and it tells him something useful: he can spend a lot more before it stops making sense.
If he wants 5 new jobs a month, that's about 17 enquiries, which is about 280 clicks, which is a budget of around $1,400. That's how you set a number. You pick the number of customers you want and let the maths price it.
Run the same sums with a worse result, say 3 enquiries per 100 clicks. Now each job costs about $555. Still worth it, but much tighter. That's the point of doing this before you spend, not after.
Why tiny budgets quietly fail
Both Google and Meta learn from results. The more wins they see, the better they get at finding people like those who already said yes. Too small a budget starves them.
On Meta, the ad system wants around 50 results per ad set (a group of ads sharing one audience and budget) in a week before it settles down. That's Meta's own guidance. If a lead costs you $30, that's $1,500 a week, or about $214 a day, for a single ad set. Below that you'll often see a "Learning limited" label in the account, and results bounce around from day to day.
You don't have to hit that figure to get value. But it explains why a $10 a day Facebook campaign feels random. It isn't broken, it just hasn't got enough to learn from.
Google is similar. The automatic bidding settings (where Google sets your price per click for you) work best with plenty of past results. Most advice puts the comfortable mark at 30 to 50 conversions a month. At $83 an enquiry that's $2,500 to $4,000 a month, which is more than many small businesses want to spend.
So if your budget is smaller, change how you set it up:
Run one campaign, not five. Every extra campaign splits a small budget into pieces too thin to learn from.
Keep your area tight. Target the suburbs or region you actually serve, so clicks aren't wasted.
Pick a simple bid setting. Under about 15 results a month, many specialists suggest setting your own price per click at first. Once you have more data, switch to "Maximize conversions", and only later add a target cost per result.
Count the right thing. A small account can't afford to chase page views. If you haven't checked what's being counted as a result, fix that first. In Google Ads, open Goals, then Conversions, and make sure only real enquiries and calls are marked as primary.
Where to put the money: Google or Meta?
The two do different jobs, and the sums above hint at which one fits.
Google catches people who are already looking. Someone types "emergency plumber Tauranga" and you show up. These clicks cost more, but the person is ready to buy. If your customers search when they have a problem, start here.
Meta puts you in front of people who weren't looking. Great for things people don't search for until they see them: a new café, a home renovation idea, a seasonal offer. Clicks are cheaper, but fewer people are ready to buy right now, so you rely on a good offer to get them moving.
A rough rule: if you can only afford one, ask whether your customers go searching or need to be shown. Search-driven jobs go to Google. Everything else, Meta. If you want the whole picture of how the two fit together, our full guide to Google and Meta ads covers setup from the ground up.
A sensible way to start
Don't commit for a year. Pick a test budget you can lose without panic, the figure your sums say is enough to produce at least a handful of enquiries, and run it for four to six weeks. Don't judge it after three days. Then look at the real cost per enquiry and per job, put those numbers back into the sums, and move the budget up or down.
Most owners find the first month tells them more than all the advice they read beforehand.
