How much to spend on Google Ads before you know if it works.
Work out the test budget backwards rather than picking a monthly figure. You need roughly thirty conversions before a result means anything, so multiply thirty by a realistic cost per lead for your industry and that is what the test costs in total. Spread it over at least six weeks, because a budget spent too quickly buys volume at bad prices and a budget spread too thin never leaves the learning period. If that number is more than you can lose, the honest answer is that paid search is not the first thing to spend on.
The question is not how much per month
Most budget advice answers a question nobody actually has. Ten percent of revenue, or whatever a competitor spends, tells you what is normal. It does not tell you what it costs to find out whether this works for you.
The useful framing is that a test has a price, the same way a piece of equipment does. You are buying an answer. The answer costs whatever it costs to gather enough data that the result is not luck, and that is a number you can calculate rather than guess.
Thirty conversions is the floor, not the target
Below roughly thirty conversions, a difference between two weeks is noise. You cannot tell a good campaign from a bad one, and neither can the bidding algorithm, which is why accounts that never reach that volume stay stuck in a permanent learning phase.
So the arithmetic is short. Take a realistic cost per lead for your industry, multiply it by thirty, and that is the cost of the answer. If leads in your market cost forty pounds, the test costs about twelve hundred pounds before you know anything. If they cost two hundred, the test costs six thousand.
That second number is the one that ends a lot of conversations honestly, and it should. A business that cannot lose six thousand pounds finding out should not be finding out with paid search first.
Spread matters as much as size
The same budget spent in two weeks and spent over eight weeks buys different things. Compressed, you outbid yourself, you catch one fortnight of demand that might have been unusual, and you never see a weekly pattern. Spread, you see which days convert, you give the bidding something to learn from, and a bad week does not end the test.
Six to eight weeks is the usual minimum for a business with a short sales cycle. If your customers take a month to decide, add that month, because otherwise you are measuring the clicks and missing the sales they turned into.
What eats the budget before it ever reaches a buyer
Three things quietly consume a test budget, and none of them are the bid.
- Broad match with nobody reading the search terms report. The account finds queries next to yours and spends real money on them, which is one of the reasons a cost per lead climbs while nothing in the setup looks wrong.
- A landing page that was never built for the search. Sending paid clicks to a homepage is the most expensive habit in small business advertising, because every extra decision between arrival and action costs conversions.
- Tracking that counts the wrong thing. If the conversion number is not the number of real enquiries, the test produces a confident answer to a question you did not ask.
Set the ceiling before you start, not after
Decide two numbers in advance. The total you are willing to spend on finding out, and the cost per lead above which this is not worth continuing. The second one comes from your own margin rather than from a benchmark, and our break even ROAS and target CPA calculator will give it to you in a minute from figures you already know.
Writing both down before the first click is the difference between a test and a subscription. Without them, month four arrives and nobody can say whether this is working, so it continues.
So what would you actually be buying
Run the arithmetic on your own business this week. What does a customer earn you, what can you afford to pay for one, and what does thirty of them cost at that price. If the total is affordable, you have a test worth running. If it is not, you have saved yourself a quarter.
If the number comes out uncomfortable and you are not sure whether the fault is the market or the setup, send us the account and the offer. We will tell you what the test should cost before you spend it, and whether paid search is even the right first place for it.
Questions we get asked.
How much should a small business spend on Google Ads per month?
- There is no correct monthly figure. Work out the total cost of the test instead: about thirty conversions at a realistic cost per lead for your industry, spread over six to eight weeks. That total, divided by the weeks, is your monthly number, and it comes from your own economics rather than from a benchmark.
Why thirty conversions?
- Below roughly that volume, the difference between two periods is usually noise rather than signal, and automated bidding has too little to learn from. It is a floor for being able to read the result at all, not a target to aim for.
Can I start with a very small daily budget?
- You can, but understand what it buys. A budget too small to reach meaningful conversion volume produces months of data nobody can act on, which costs more in time than a shorter properly funded test would have cost in money.
Should I pause the campaign if the first two weeks look bad?
- Almost never. The first weeks include the learning period, and pausing restarts it. Set the total you are willing to spend and the cost per lead you cannot exceed before you launch, then hold to both rather than reacting weekly.
What if my industry has very expensive clicks?
- Then the test costs more, and that is worth knowing in advance rather than discovering halfway through. In expensive markets it is often better to fix conversion rate on the page first, because every point of improvement there lowers the cost of every test that follows.
