There's no single right Google Ads budget, and a number quoted before anyone asks about your business is a guess. The right amount is enough to win a customer for less than that customer is worth to you, with a bit of room to learn what works. So start with what a customer is worth, then run a test you can afford to lose.
Four things move the number: competition, location, your margins and your goal. Here's how to work out your own figure, how to test it, what counts as enough data, and when to stop.
What drives the number
Competition
Google sells ads in an auction. When lots of businesses want the same search, each click usually costs more. So the same budget buys fewer clicks in a busy market.
In plain English: CPC means cost per click. It's what you pay each time someone clicks your ad. It's an average, and it changes by search term, area and season.
Location
Serving a few suburbs is a different auction from covering all of Sydney. A smaller area usually means fewer searches, so a smaller budget can be enough. A wider area usually needs more.
Margins
If you keep a good share of each sale, you can afford to pay more to win a customer. If your margin is thin, you can't. Margin is what's left after the costs of doing the job.
Your goal
Be specific. A few extra calls a week needs a different budget from a full diary. And ads only help if you can take on the work. If you're already flat out, the right budget might be zero.
Work backwards from what a customer is worth
Start with the customer, not the click. These five steps give you a ceiling, which is the most you can sensibly spend. There's a bit of maths, but nothing that won't fit on the back of an envelope.
1. Work out what a new customer is worth to you. Use the profit from the first job, and add repeat work if that's normal for you. 2. Decide how much of that you're happy to spend to win them. That's your ceiling for one customer. 3. Check how many enquiries turn into customers. Look at your last batch and count. If you don't track this, start now. 4. Shrink the ceiling down to one enquiry. Multiply your ceiling per customer by the share of enquiries that become customers. 5. Multiply that by the enquiries you want each month. That gives you a monthly budget to test against.
A worked example:
If the budget you can afford is far below what the clicks are likely to cost, the maths might not work. Google shows rough estimates for your search terms inside Google Ads, so you can find that out before you spend a cent.
Ad spend is what you pay Google for clicks. Management is a separate cost for whoever runs the account. My pricing ranges show the management side, so you can add the two together.
What this looked like for one real business:
A test-budget approach
A test is a small, time-limited experiment with one question. Do people in your area who search for this turn into enquiries you can win, at a cost you can afford? Start small. I'd rather you learn something cheaply than find out the expensive way.
- Pick an amount you can lose. It shouldn't hurt your cash flow if it brings in nothing. Any good agency will agree the figure with you in writing before anything goes live. My minimum is a month.
- Start narrow. Pick one service, one area, and the searches closest to buying. Add more only when something works.
- Set the length first. Google Ads usually takes a few weeks before the numbers mean much. Pick the end date before you start, so you aren't judging on day three.
- Track enquiries, not clicks. Count calls and form fills. If calls aren't tracked, you're guessing.
- Change one thing at a time. Otherwise you can't tell what made the difference.
How one test went:
What "enough data" means
Enough data means enough clicks and enquiries that one lucky or unlucky week wouldn't change your mind. A handful of clicks tells you almost nothing. Two calls in one day tell you nothing either.
Judge the test on enquiries and customers. Clicks and impressions are only the first step. Look at the trend over several weeks, and compare like with like. A quiet season can look worse than a busy one even when the ads are doing fine.
My rule of thumb:
Honest note: A small budget can still work, but it learns slowly. You may need to narrow the area or the service to collect enough data to judge. Sometimes the honest answer is to wait, or to fix your website or offer first.
Signs to stop or pause
Stopping an ad test isn't failing. It means the test did its job. Pause when any of these is true:
- You've run the test for the length you set, with enough data, and each enquiry costs more than it's worth to you.
- The enquiries are the wrong kind, like people outside your area or people who only want the cheapest price, and changes haven't helped.
- You can't answer calls or follow up quickly. Fix that first.
- Tracking is broken, so you can't tell what's working.
- You can't take on more work right now.
- Nobody's explaining the numbers to you in plain words.
A time I said pause:
Where to go next
If you want to see how I'd approach it, read about my Google Ads management. Choosing between channels? See Meta ads vs Google Ads.
When you're ready, let's chat. I read every message myself, and if ads aren't right for you yet, I'll say so.