How to Tell If Your Google Ads Are Wasting Money

The question your dashboard cannot answer

Most ad reporting shows impressions, clicks, and a cost per click, and all three will usually look fine. They are also close to meaningless on their own, because none of them tell you the only thing that matters: how many customers this produced, and what each one cost.

If you have asked that question and been shown a chart rather than given a number, that is the answer. Nobody knows.

Failure one: you are paying for searches you would reject on sight

Google does not only show your ad for the keywords you chose. Under broad match — which is the default, and which Google actively encourages — it shows your ad for anything it judges related.

In practice that means an emergency plumber pays for people searching how to fix a dripping tap themselves. A commercial roofer pays for homeowners pricing a shed. A dental implant clinic pays for students researching implant materials for coursework. The clicks are real, the budget is really spent, and not one of those people was ever going to buy.

This is visible, for free, in about ten minutes. In Google Ads, open the Search Terms report. That is the list of what people actually typed, as opposed to the keywords you bid on. Read it. Most owners find the waste in the first thirty rows, and it is usually obvious enough to be annoying.

Failure two: nothing is counting the outcomes

This is the more expensive failure, and it is quieter.

If the account cannot see which clicks became enquiries, then Google's automated bidding is optimising toward whatever goal it was given — usually clicks. It will get you clicks. It will get you cheaper clicks over time. It has no idea whether any of them turned into work, because nobody told it.

And neither do you. Which means you cannot answer whether to spend more, spend less, or stop. In the absence of an answer, almost everybody keeps paying the same amount indefinitely, because that feels like the safe option. It is not. It is just the option that requires no decision.

The order of operations matters

There is a strong temptation to start by rewriting ads or adding keywords. That is the wrong first move, because you will not be able to tell whether it helped.

First, conversion tracking. A few hours of work. Form submissions and phone calls recorded as events, tested with a real submission rather than assumed to work. Until this exists, every other number in the account is decoration.

Second, the search terms report. Add negative keywords for the obvious waste. Tighten match types where broad is clearly not earning its place.

Third, exclusions. Placements, audiences and locations that spend without producing. The Display Network in particular will happily spend a budget on apps you would never choose.

Only then, scale. Once you know what an enquiry costs and that number is stable, increasing budget becomes a decision with a predicted outcome rather than a hope.

Expect the first month to cost less, not more

This surprises people who expect an agency to ask for a bigger budget. The first month of this work usually reduces spend, because most of it is subtraction: cutting searches that were never going to convert, and stopping placements that were quietly consuming budget.

That is also the honest test of whether somebody is working for you. An account that only ever grows, with no month where spend came down, is an account nobody has looked at closely.

What good looks like after sixty days

You should be able to say, without checking with anybody: this is what an enquiry costs us, this is roughly how many we get a month, and this is what happens if we add another thousand dollars.

If you cannot say those three things, the account is not being managed. It is being maintained.

What a well-run account looks like from the outside

You do not need to understand the mechanics to tell whether somebody is doing this properly. Three questions will do it.

Ask what your cost per enquiry is. A managed account has this number and it is stable enough to quote. If the answer is a cost per click, or a chart, the outcomes are not being counted.

Ask what was turned off last month. Every healthy account has subtractions — negative keywords added, a placement excluded, a poorly performing ad paused. An account that only ever grows is an account nobody is reading.

Ask which search terms you paid for that you would not have chosen. Anybody managing your account seriously has read that report recently and will have examples, usually with some irritation.

The trap of judging on cost per click

Cost per click is the number most reports lead with and it is close to useless in isolation. It can always be reduced by bidding on cheaper, less relevant searches — and cheaper searches are cheaper precisely because the people making them are further from buying.

An account can halve its cost per click and produce fewer customers, and the report will look like an improvement. This is why conversion tracking is not an optional refinement: without it, the metric everybody optimises toward actively rewards attracting the wrong people.

Related: Your ad spend is unaccountable