Skip to main content
Clicks & ClientsClicks&Clients
08 / Field Notes
Google AdsJuly 31, 20266 min read

How much should I spend on Google Ads?

How much should I spend on Google Ads? Set a floor rather than a percentage of revenue. Under about $1,000 a month in media, most trades should skip it.

Most budgeting guides answer this with a percentage of revenue, usually 5 to 10%. That rule is fine for a marketing plan and useless here, because the auction has no idea what your revenue is. The number that decides whether this channel works for you is a floor: a monthly media budget below which you buy too few clicks to learn anything, and above which the account starts telling you the truth about your market. In most trades that floor sits near $1,000 a month. In law it is closer to three times that.

Where the floor comes from

The floor exists because Google charges by the click and pays you nothing until a click turns into a ringing phone. Across 13,474 US search campaigns measured from April 2025 through March 2026, the average click cost $5.42 and the average lead cost $66.69 1. Divide the second by the first and it takes about twelve clicks to produce one call or form fill.

So $1,000 a month buys roughly 185 clicks and about fifteen leads. Fifteen is a thin month. It is enough to tell a good month from a disastrous one and not enough to tell a good month from an ordinary one, which means at that level you are funding an education rather than a lead flow. Below it things fall apart fast. At $500 you get seven or eight inquiries, two of which will be a wrong number and somebody asking whether you rent equipment.

Build the number up from one job

Start at the far end, with a booked job, and walk backward until you arrive at a budget. Owners ask how much should I spend on Google Ads as though the number is sitting in a benchmark report somewhere. It is sitting in your close rate.

Say a job bills $600 and leaves you $240 after parts and labor, and you win one quote in four. Four leads at the $66.69 average cost $267, so that booked job cost more to buy than it earned. No budget fixes that. The advertising is doing its job and the loss is happening between the ringing phone and the signed invoice, which is where most disappointment in this channel lives.

Now change one assumption. Same job, same $240 of profit, but you win one quote in two. Two leads cost $133 and the job clears $107. Thin, but positive, and it repeats every month you fund it. At that close rate, six new jobs a month needs twelve leads, or about $800 in media, which lands under the floor. You fund the floor anyway and take seven or eight jobs instead of six, because the auction does not sell a smaller version of itself.

Everything in that example turns on the close rate, which is yours to know and nobody else’s to guess. Our published benchmark bands include a typical close rate for twenty-odd trades, and you can run your own version of this in the lead cost calculator in about two minutes.

Under $1,000 a month, we turn the work down

Here is the rule we apply to our own pipeline. If a business in a standard trade can put less than about $1,000 a month into media, we tell them not to run Google Ads at all, and we decline the account rather than bill a management fee on top of a budget that cannot work. Not “start small and prove it out.” Don’t start.

The reason is mechanical. Fifteen leads a month, split across four services and three towns, leaves you two or three per line item, and no bidding system, automated or human, makes a decent decision on two data points. You pay full auction price for the privilege of being a rounding error in somebody else’s market.

There is some evidence that owners already sense this. Fifty-two percent of small businesses run monthly marketing budgets under $1,000, and among that group only 30% run search advertising at all 2. The majority sitting it out are making a defensible call.

If you are under the floor, the money has two better homes. The first is the calls you already get and lose, which costs nothing but attention. The second is a pay-per-lead format where you are billed when the phone rings rather than when someone clicks, which caps the downside that kills small budgets. Our ad spend calculator will run your trade’s numbers and give you a verdict, and the verdict is sometimes no.

Law firms need a bigger floor than repair shops

Legal is the most expensive category in the 2026 benchmarks at $131.63 per lead, against $29.96 for automotive repair 1. The same $1,000 buys a law firm seven or eight inquiries and a repair shop thirty-three. Seven inquiries is an anecdote, not a month of data.

So the floor scales with what a lead costs where you compete. A family law practice or a defense firm should treat $3,000 a month in media as the entry price, and a personal injury firm bidding against the billboards on I-85 should double that again before expecting anything readable. A detail shop or an oil change business clears the floor at $600. Same channel, same math, wildly different entry fee, which is most of what we spend the first call explaining when someone asks about running ads in Atlanta.

2026 is the first year in five that the price moved your way

Cost per lead fell year over year for the first time in five years, and conversion rates rose in 87% of industries 1. In plain terms, the same budget now produces more calls than it did last year across most of the economy.

That is an argument for holding a budget steady rather than pulsing it. An account that runs at the floor for a full quarter produces about forty-five leads, which is a sample you can reason about. The same money spent in three disconnected bursts produces three unreadable months and a conclusion that Google Ads doesn’t work in your market. We wrote separately about what Google Ads cost once an account is running, and the pattern holds there too.

Three things people push back on

“Can’t I start at $300 and work my way up?”

You can, and you will spend four months learning what one month at the floor would teach. The clicks are not cheaper at $300; there are just fewer of them, and the ones you get are spread thinner. If $300 is genuinely what is available, spend it on answering the calls you already miss.

“What about $50 a day?”

Fifty dollars a day is $1,500 a month, which clears the floor in every trade except law and a handful of high-cost categories. That is a workable starting budget if you point all of it at one service in one geography and leave it alone for ninety days.

“Does the floor include what I pay someone to manage it?”

No. The floor is media only, and media goes from your card straight to Google with nobody in between taking a cut. Management fees sit on top, ours start at $1,850 a month and are published on our services page, and that stacking is exactly why we send people under the floor away instead of onboarding them.

Which leaves the number this piece opened with sitting there unresolved. Just over half of small businesses run on marketing budgets under $1,000 a month, and 70% of that group never runs a search ad 2. We cannot tell from the data how many of them priced a lead in their own trade and walked away, and how many never asked. Our read is that the second group is much larger, and that most of them would come out under the floor anyway.

Sources
  1. 1.WordStream: Google Ads Benchmarks 2026 · accessed 2026-07-31
  2. 2.LocaliQ: Big Small Business Marketing Trends Report 2026 · accessed 2026-07-31
From the firm

Field Notes is the public version of the working theory we run on every account. If you want to talk about your own, book a discovery call.

More from Field Notes
See all →