Insights

How to set a Google Ads budget when you are starting out

Start from what a new customer is worth, work back to a daily budget you can afford to test, and understand what Google actually does with that number.

By the Green Branch team · · 7 min read

Tablet on a wooden surface showing a web analytics dashboard

"How much should I spend on Google Ads?" is usually the first question we hear. The honest answer is that the right number comes from your own business, not an industry average. A budget picked because it feels comfortable is often too small to learn anything, or larger than the results can justify.

The better approach is to work backwards from what a customer is worth. It takes a few minutes with a calculator, and it also tells you whether paid search is likely to work for you before you spend anything.

Start with what a customer is worth

Begin with the value of a typical new customer. Use gross profit rather than revenue, because money needed for materials and wages cannot also be spent on advertising.

Decide over what period you will count it. If customers usually buy once, use that one sale. If they come back, a first-year figure is a sensible middle ground. Lifetime value is tempting, but early on it is mostly an optimistic guess.

Then decide how much of that profit you are willing to spend to win the customer. That figure is your affordable cost per acquisition, often shortened to CPA. It is a business decision, not a marketing one.

Work backwards to an affordable cost per click

From the affordable CPA, three more numbers get you to a daily budget:

  1. Close rate: what share of inquiries become paying customers. If you close one lead in four, you can afford a quarter of your CPA for each lead.
  2. Conversion rate: what share of ad clicks become inquiries. Google defines it as conversions divided by ad interactions, so 50 conversions from 1,000 interactions is 5%. Multiply your affordable cost per lead by this rate to get an affordable cost per click.
  3. Clicks needed for a fair test: enough clicks that you can expect a meaningful number of leads, divided across the days you will run the test.

If you have never advertised, you will not know your conversion rate yet. Use a cautious estimate and replace it with real data once the campaign has been running for a few weeks.

A worked example with hypothetical numbers

Hypothetical example. The business and every figure below are invented to show the arithmetic. They are not benchmarks, and your own numbers will be different.

Imagine a home cleaning company serving Ann Arbor. A typical new customer brings in $1,200 in their first year at a 50% gross margin, and the owner is prepared to spend up to a third of the gross profit to win one.

StepHypothetical inputWorkingResult
Value of a new customer$1,200 first-year revenue at 50% margin$1,200 × 50%$600 gross profit
Affordable cost per customerSpend up to one third of gross profit$600 ÷ 3$200
Affordable cost per lead1 in 4 leads becomes a customer (25%)$200 × 25%$50
Affordable cost per click1 in 20 clicks becomes a lead (5%)$50 × 5%$2.50
Clicks for a first readAim for about 15 leads15 ÷ 5%300 clicks
Daily budgetSpread over 30 days: 10 clicks a day10 × $2.50$25 a day

Fifteen leads is enough for a first read, not a final verdict. It will tell you whether the phone is ringing and whether the inquiries are any good.

Next, compare the affordable cost per click with what your keywords tend to cost. In Google's Keyword Planner, the top of page bid ranges approximate the 20th and 80th percentiles of what advertisers have historically paid for a top-of-page position. They are historical figures rather than a forecast, but if they sit far above your affordable cost per click, the arithmetic is telling you something. You can improve the landing page to lift the conversion rate, focus on more specific searches, raise the value of each sale, or decide that another channel is a better place to start.

How Google Ads daily budgets actually behave

Most Google Ads campaigns run on an average daily budget, and the word "average" matters. Google's overdelivery rules work like this for most campaigns:

  • On any single day, a campaign may spend up to two times its average daily budget, to take advantage of days with more search traffic.
  • Over a month, you will not be charged more than 30.4 times the average daily budget, 30.4 being the average number of days in a month.

So the $25 a day in our example could cost up to $50 on a busy day, but no more than $760 across a month. Google states that you will never pay more than these daily and monthly spending limits. If you change the budget partway through a month, the rest of the month paces to the new daily budget multiplied by the remaining calendar days.

If your average daily budget is lower than the amount Google recommends, the campaign may show a "Limited by budget" status, and Google shows your ads less often so the budget lasts. That is not necessarily a problem during a test, as long as the clicks you do get are good ones.

Set up conversion tracking before you spend

Without conversion measurement, you can see clicks and costs but not which ones produced customers. Every calculation above depends on knowing your cost per lead, so set tracking up first and test it.

  • Forms: track the thank-you page or the form submission as a conversion.
  • Calls: Google Ads can track phone calls from your ads and to a number on your website, counting calls that last at least a minimum length you set.
  • Sales that happen offline: if leads turn into sales over the phone or in person, Google supports offline conversion imports and recommends starting with enhanced conversions for leads. Feeding back which leads became customers is how you measure quality as well as quantity.

Before launch, submit a test inquiry and make a test call, and check that both are recorded.

Start narrow, then widen

A small budget goes furthest when it is concentrated on the people most likely to buy.

Keywords and match types

Google offers three keyword match types: broad match (the default), phrase match written in quotes such as "house cleaning", and exact match written in square brackets such as [house cleaning service]. Google recommends pairing broad match with Smart Bidding, and that can work well once an account has reliable conversion data. While tracking is new and the budget is tight, we usually prefer phrase and exact match on a short list of high-intent searches, so spend is easier to control and read. Broad match is worth testing once you have a steady record of conversions.

Location

Target only the area you serve. A business covering Ann Arbor and a few neighboring towns gains nothing from showing ads across the whole state. Also check the location options: the default reaches people in or regularly in your area and people who have shown interest in it. Google suggests the "Presence" option when you only want to reach people in your chosen locations, which is often the right choice for a local service.

Networks

Google includes search partners by default in Search campaigns, and you can opt out in campaign settings. On a small test budget, we generally start with Google Search alone so the results are easier to interpret, then test partners separately.

Negative keywords

Negative keywords stop your ads showing for searches you do not want, such as "jobs", "free" or "DIY" for most service businesses. Negative keywords do not match close variants, so add plurals and synonyms yourself. Check the search terms report every week at first, and add irrelevant searches as negatives.

Allow time for learning

If you use an automated bid strategy, Google may show a "Learning" status after you create the strategy or change its settings, keywords or campaigns. Google explains that the learning period depends on how many conversions you get, how long your conversion cycle is, and which bid strategy you use, and that it typically takes one to two conversion cycles to calibrate.

In practice, that means two things. Judge results over weeks, not days. And avoid making large, frequent changes, because each change to an automated strategy can trigger a new learning period. Batch your changes and give each batch time to show its effect.

When to raise the budget

Increase spending when all of these are true:

  • Conversion tracking has been checked and you trust the numbers.
  • Your actual cost per lead has been at or below your affordable figure over a meaningful number of leads, not just a good week.
  • The leads are turning into customers at roughly the close rate you assumed.
  • The campaign is limited by budget, so there are more searches available to buy.

Raise the budget in steps rather than all at once, and watch the cost per lead after each step. Extra budget often buys less efficient clicks at the margin, so the question is whether the extra customers are still worth what they cost.

If the numbers are not working after a fair test, look at the landing page, the offer and the search terms before writing the channel off. A weak page often explains a high cost per lead better than the bids do.

Want help setting up a first campaign?

If you would like someone to check the arithmetic, set up tracking or run the first test with you, that is what our pay-per-click advertising service is for. We build campaigns around cost per lead, not clicks. Get in touch and tell us what a new customer is worth to you.

Frequently asked questions

Is there a minimum spend for Google Ads?

No. Google says there is no minimum spending commitment and that you set and control your own budget. The practical minimum is the amount that buys enough clicks to learn something within a reasonable time, which is why we suggest working it out from your own numbers.

How do I turn a monthly budget into a daily one?

Divide the monthly amount by 30.4, the average number of days in a month that Google uses. Google's own example is a $304 monthly budget, which becomes $10 a day. Remember that spending on individual days will vary around that average.

How long should a first test run?

Long enough to collect a meaningful number of leads and to get past any learning period, which for most small local campaigns means several weeks rather than days. If your customers take a while to decide, allow longer, because sales from early clicks may not arrive until later.

What if I do not know my conversion rate yet?

Start with a cautious estimate, run the test, and replace the estimate with your real figure as soon as you have enough data. The point of the calculation is not to be precise on day one, it is to make sure you know which number to check and what it needs to be.

Keep reading

More insights

Ready to grow?

Tell us about your business and what you want marketing to achieve. We will come back with honest advice on where to start.

Start a project