How Much Should a Small Business Spend on Google Ads?

It is usually the first question a business owner asks about Google Ads, and it almost always gets a round-number answer: what feels affordable, what a competitor seems to spend, what an industry article calls typical. Round numbers are how budgets go wrong. The number gets picked before anyone asks what the money is supposed to do.

There is a better way to arrive at it. Call it budgeting backward: instead of choosing a monthly figure and hoping, you start from what a new customer is worth to your business, work back through what you can afford to pay for one, and let the budget fall out of the math. The result is a number you can defend, because every part of it comes from your own economics rather than someone else's average.

The second thing this post argues is less obvious: for many accounts, the right first move is not raising the budget at all. But start with the math.

Start from what a customer is worth

Budgeting backward has three steps, and none of them requires a spreadsheet.

First, put a value on a new customer. For some businesses that is the first sale. For businesses where customers come back, like a salon, a clinic, or a service on subscription, it is closer to what a customer spends over a year. Use the honest figure, not the hopeful one.

Second, decide what you would gladly pay to acquire one customer at that value. This is a margin decision only you can make. Pay too close to the full value and you are buying revenue, not profit. Leave room, and every customer the ads bring in makes the business stronger.

Third, multiply that acceptable cost per customer by the number of new customers you want in a month and can actually serve. Capacity is part of the math. There is no point paying for demand your calendar cannot absorb.

That product is your budget. Not a figure from an article, but the direct consequence of what a customer is worth to you, what you will pay for one, and how many you can handle. When any of those three inputs changes, the budget changes with it, and you know exactly why.

The floor: enough data to learn

There is one constraint the backward math has to respect. Google's auction, and anyone managing your account, learns from outcomes. Clicks that convert teach the account what a good search looks like. Clicks that do not convert teach it what to stop paying for. A budget too small to produce a steady flow of clicks produces almost no lessons, so the account never improves. It just spends slowly.

So the practical floor is not a dollar amount. It is a data amount: enough activity each day that a weak keyword reveals itself within weeks, not quarters.

If your backward math lands below that floor, the fix is not to thin the same plan across less money. It is to narrow the plan until the budget is concentrated again: fewer keywords, a tighter service area, one flagship service instead of five. A small budget focused on one thing generates real evidence. The same budget spread across everything generates noise.

Before you raise it, fix where it flows

Here is the part most budget advice skips. A monthly budget tells you how much money enters the account. It tells you nothing about how much of that money reaches searches that turn into customers. On a neglected account, a meaningful share of the budget leaks into searches that never convert, which makes the budget look too small when it is actually misdirected.

That is why "should I spend more?" is usually the wrong first question. The right one is "where is the current spend going?"

A real account we manage makes the point. An escape-room business in the US was already profitable when we took it on. Instead of raising the budget, the work went into where the money flowed: cutting the searches that spent without booking, moving budget toward the campaigns that earned it. Spend fell 36 percent. Return on ad spend rose from 2.8× to 3.8×. Cost per booking dropped 17 percent. Less money in, more value out, because the dollars stopped subsidizing searches that never converted.

The order matters. Fix the flow first, and every additional dollar you add later has somewhere productive to go. Raise the budget first, and you scale the leaks along with the wins.

How the AI managers handle your budget

This is the job AdvisorPPC's AI managers do daily. The Auditor reads your account and never writes to it: it shows you where the current budget actually flows, which searches convert, and which ones only spend. That audit is how you find out whether your budget problem is size or direction before you commit another dollar.

Then the Optimizer proposes the moves: budget shifted from a campaign that stopped converting to one that is winning, negative keywords to stop paying for searches that never buy, bids trimmed where the price no longer makes sense. Each proposal is shown to you in plain English with its expected impact.

Nothing runs until you approve it. No spend, bid, or budget change happens without your sign-off, and every change that runs is logged with the reason behind it. The monthly cap stays yours: the AI managers move money within the budget you set, and never past it. Daily corrections, inside your ceiling, with your yes on every move.

A number you can defend

The honest answer to "how much should I spend?" is a short chain of your own facts: what a customer is worth, what you will pay for one, how many you can serve, and whether your current spend is reaching searches that convert.

So, in order: do the backward math and write the number down. Check it clears the data floor, and narrow the plan if it does not. Run the read-only audit before you raise anything, and cut the waste it finds first. Then, and only then, scale. Do it once the account has shown, the way 2.8× became 3.8× on less spend, that additional dollars have somewhere good to go.

The budget question never really goes away; you will revisit it as prices shift and seasons turn. But budgeting backward turns it from a guess you renew every month into arithmetic you check, and every change to it stays a change you approved.

AdvisorPPC's AI managers are built using the Claude API from Anthropic. AdvisorPPC is not affiliated with or endorsed by Anthropic.