How much should you spend on Google Ads?

The question is usually asked the wrong way round. People ask what they should spend, when the only answerable question is what a customer is worth to them and how many they want next month.

Work backward from a customer, not forward from a budget

You need three numbers. Most business owners can produce all three in about ten minutes, and the exercise is worth doing even if you never run an ad.

1. What a customer is worth. Not the first invoice, the whole relationship. A dentist should count the crown, the check-ups and the family members who follow. A B2B service should count the contract value times the average number of renewals.
2. What share of enquiries become customers. Ask the person who answers the phone rather than guessing. Twenty to forty percent is normal for a service business with a decent follow-up habit, and considerably lower without one.
3. What you are willing to pay to acquire one. Some fraction of the first number, sized to your margin and how patient your cash flow is.

Now the arithmetic. If a customer is worth 3,000 and you are willing to spend 300 to win one, and one in four enquiries closes, then you can pay up to 75 for an enquiry. If your category's clicks cost 5 and your landing page converts at 6 percent, an enquiry costs you about 83, and the account does not work yet. That is not a reason to abandon it. It is a specific instruction: raise the conversion rate, lower the click cost through tighter targeting, or improve the close rate. Each of those is a fixable problem, and you now know which one to attack.

The minimum budget that lets an account learn

Separately from what you want to spend, there is a floor below which a Google Ads account cannot function. It has nothing to do with ambition and everything to do with data volume.

Smart bidding needs conversions to learn from. A rough working threshold is around 30 conversions a month per campaign before automated bidding behaves predictably. Below that, the system is guessing, and so is anyone reading the reports.

As a practical floor, aim for enough budget to buy roughly 30 clicks a day on your main terms. Multiply your category's typical click cost by 30, then by 30 days, and that is your realistic minimum monthly spend. In a cheap category that lands around a few hundred a month. In legal, insurance or B2B software, it lands in the thousands.

If that number is beyond you right now, do not run a thin account across everything. Narrow instead: fewer keywords, one city, one service, tighter match types. A small budget concentrated on your single most valuable search can work. The same budget spread across forty keywords buys you nothing but a report.

What clicks actually cost by category

Click costs vary enormously and the variation is not random. It tracks what a customer is worth, because your competitors have done the same arithmetic you just did.

Ranges published online are averages across countries and seasons, and your auction is local. The only cost that matters is the one in your own market, which the keyword planner will show you in a few minutes, and which the first fortnight of live spend will confirm properly.

Splitting the budget once you have one

For a service business starting fresh, a structure that works in most accounts:

• Around 70 percent on high-intent search. The terms where somebody is describing the exact problem you solve, often with a location or an urgency word attached. This is where the leads are.
• Around 20 percent on secondary and exploratory terms. Related services, competitor terms if the economics allow, and the queries you suspect are valuable but have not proven yet. This is your research budget and it should be deliberate.
• Around 10 percent on remarketing. Cheap, and it recovers people who were interrupted rather than uninterested.
• Brand terms: only if someone is bidding on your name. Otherwise you are paying for traffic you were about to get free. Check monthly, because a competitor can start any week.

Performance Max sits outside this until search has proven the economics and your conversion data is clean. Launched too early it absorbs brand traffic, reports it as new demand, and makes a mediocre account look excellent.

Is your current spend buying leads or buying data?

Four checks you can run today on an existing account. Any one of them failing means budget is leaking.

1. Open the search terms report for the last 30 days. Read it line by line. If more than a fifth of the spend went to queries you would never have chosen, your negative keyword work has stopped.
2. Count your conversion actions. If page views, newsletter signups and five second phone calls are all counted as conversions, the bidding algorithm has been trained on noise and is buying accordingly.
3. Separate brand from non-brand. Look at cost per lead with brand terms excluded. This is often the moment an account's real performance becomes visible for the first time.
4. Check where each campaign lands. If everything points at your homepage, you are paying twice: once in a lower conversion rate and again in a worse quality score that raises every click.

When the honest answer is to spend nothing

There are situations where the right recommendation is to keep your money, and an agency paid a percentage of your spend will rarely make it.

• Nobody is searching for what you sell. If the volume is not there, ads cannot create it. That is a Meta Ads or content problem instead.
• Your close rate is the actual constraint. If half your enquiries never get called back, buying more of them is expensive. Fix the follow-up first and the same budget produces more customers.
• Your landing page converts at under one percent. Doubling the conversion rate is almost always cheaper and faster than halving the click cost.
• You cannot sustain three months. The first month is mostly waste removal. Judging an account after four weeks and stopping is the most common way businesses lose money on Google Ads.