The Hour
You can find most of what's wrong with a Google Ads account in sixty minutes, without any tools beyond the account itself. Here's the order we go in, and roughly how long each step takes.
- Search terms report — 15 minutes
- Conversion tracking — 10 minutes
- Campaign and match type structure — 10 minutes
- Quality Score and ad relevance — 10 minutes
- Landing pages, 10 minutes
- Budget allocation and wasted spend, 5 minutes
We're giving this away in full because the businesses that run it and find nothing wrong were never going to hire us, and the ones that find four things wrong usually want help with the fifth.
1. The Search Terms Report (Start Here, Always)
This is where the money is, and most account owners have never opened it.
Go to Campaigns, then Insights and reports, then Search terms. Set the date range to the last 90 days. Sort by cost, descending.
Now read the list. Not the keywords you bid on. The actual phrases people typed.
What you're looking for: anything you'd never want to pay for. Job seekers ("google ads jobs", "plumber apprenticeship"). Students ("google ads course", "how to become an electrician"). Free-seekers ("free", "diy", "cheap"). Competitors' brand names. Wrong locations. Wrong intent entirely.
Add every one of them as a negative keyword. Campaign level if it's specific, account level if it's universal.
We did exactly this on one account and cost per enquiry dropped by roughly a third within a fortnight, with no other change. Nothing else in this audit produces that kind of return that quickly.
The thing people get wrong: doing it once. Search terms shift continuously, especially with broad match. Set a calendar reminder for monthly.
2. Conversion Tracking (The One That Invalidates Everything Else)
If this is broken, every other number in the account is fiction and so is Google's bidding.
Go to Goals, then Conversions, then Summary. Check three things.
What's being counted? If your conversion actions include page views, time on site, or "all website visits", you're optimising for browsing. Conversions should be calls, form submissions and bookings: things a customer did.
Is anything double-counting? A common failure: a form submission counted both by a thank-you page view and by a click event. Your conversion numbers look twice as good as reality and your cost per acquisition looks half as bad.
Are conversions recorded in the last week? Zero recent conversions on an active account usually means the tracking broke, often during a website update that nobody connected to the ads.
Then the harder question: do you know which conversions became revenue? Fifteen form fills a month tells you little if twelve were tyre-kickers. Importing offline conversions from your CRM is the single biggest improvement available to most service-business accounts, and almost nobody does it.
3. Structure and Match Types
Open the keyword list. Look at the match type column.
If everything is broad match, that's usually a problem in a small Australian account. Broad match needs volume and a strong conversion signal to work well, and most Melbourne service accounts have neither. Phrase and exact give you control at the scale you're operating at.
If everything is exact match, you're probably missing volume. A reasonable mix is exact and phrase for the terms you know convert, with a tightly negatived broad campaign to discover new ones.
Check campaign separation. Search and Display should never share a campaign. Display clicks are vastly cheaper and will swallow the budget while converting at a fraction of the rate. If you see "Search Network with Display Select" anywhere, split it.
Check geography. Campaign settings, Locations. The default setting targets people "in or regularly in" your area, which is usually right. The alternative (people who've shown interest in your area) reaches people in Perth researching Melbourne. Also check you haven't excluded nothing: if you only service the north, Frankston clicks are waste.
4. Quality Score, Read Properly
Add the Quality Score column to your keyword view, along with the three component columns: expected click-through rate, ad relevance and landing page experience.
Quality Score itself is a diagnostic, not a target. The components tell you what to fix.
Low ad relevance means your ad text doesn't match the keyword. Someone searching "hot water system replacement" should see an ad about hot water system replacement, not "Melbourne's Trusted Plumbers".
Low expected CTR usually means the same problem, or an offer nobody wants.
Low landing page experience means the page doesn't match the search, is slow, or both. This is the most expensive of the three to ignore, because it affects both your cost per click and your conversion rate.
Anything scoring 4 or below with meaningful spend behind it is worth an hour of attention. Anything at 3 or below is costing you a premium on every click.
5. Landing Pages
Click your own ads. All of them. On a phone.
Does the page say what the ad said? An ad for emergency plumbing that lands on a homepage listing twelve services has lost the thread, and so has the visitor.
Is there a phone number visible without scrolling? For service businesses, on mobile, this is the single most common conversion failure we find.
Does the page load in under three seconds on 4G? Not on your office wifi. Run it through PageSpeed Insights on mobile.
Is the form short? Every field costs you submissions. Name, phone, and one line about the job is usually enough. Asking for a postcode, a preferred contact time and a budget range before you've spoken to anyone is asking people to leave.
We've taken over accounts where the previous agency spent three months refining bid strategies on a page with no phone number above the fold. The bidding was fine. What actually improves conversions goes through the rest.
6. Where the Budget Is Going
Sort campaigns by cost over 90 days. Then sort by conversions.
If your highest-spending campaign isn't among your highest-converting, you've found your problem in about forty seconds.
Two patterns we see constantly. Brand campaigns eating budget, bidding on your own business name looks efficient because the cost per conversion is low, but a large share of those people would have found you anyway. It's defensible when competitors bid on your name, and it shouldn't be your biggest line item.
Performance Max absorbing everything. PMax will take as much budget as you give it and the reporting is deliberately opaque. If it's running alongside Search campaigns with no brand exclusions, it's likely claiming conversions that Search would have delivered more cheaply. Add brand exclusions and compare.
What You Can't Audit Yourself
Being straight about the limits.
You can't easily tell whether your cost per lead is good without a benchmark for your category, and benchmarks are hard to come by honestly. You can't see what competitors are bidding. And auction-time data (the conditions under which you won or lost) needs more than an hour.
But the six checks above find the majority of wasted spend in the majority of accounts, and none of them need an agency.
If you'd like a second pair of eyes on what you find, we'll look at an account as part of the free audit and tell you what we'd change, whether or not you work with us. Our Google Ads page covers how we manage accounts, and what a click costs in Melbourne sets out the local numbers.
Frequently Asked Questions
How do I audit my own Google Ads account?
Work through six checks in about an hour: read the search terms report for the last 90 days and add negatives, verify conversion tracking counts real actions without double-counting, review match types and campaign separation, check Quality Score components, click your own ads on a phone to test the landing pages, and compare spend against conversions by campaign.
What should I look for in the Google Ads search terms report?
Phrases you'd never pay for. Job seekers, students looking for courses, people searching "free" or "DIY", competitor brand names, wrong locations, and anything with the wrong intent. Sort by cost over 90 days so the expensive mistakes surface first, then add each one as a negative keyword at campaign or account level.
How do I know if my Google Ads conversion tracking is broken?
Check three things in Goals, then Conversions. Confirm the actions being counted are real outcomes like calls and form submissions rather than page views or time on site. Look for double-counting, where one form submission registers twice. And confirm conversions have been recorded in the last seven days. Zero on an active account usually means tracking broke.
What's a good Quality Score in Google Ads?
Treat it as a diagnostic rather than a target. The useful information is in the three components: expected click-through rate, ad relevance and landing page experience. Any keyword scoring 4 or below with meaningful spend behind it deserves attention, and 3 or below means you're paying a premium on every click for that term.
Should I use broad match in a small Australian Google Ads account?
Usually not as your primary match type. Broad match needs high volume and a strong conversion signal to perform, and most Melbourne service accounts have neither. A workable structure is exact and phrase match on terms you know convert, plus one tightly negatived broad campaign used purely to discover new search terms.
How often should I check the search terms report?
Monthly, at minimum. Search terms shift continuously as Google's matching evolves and as seasonal language changes, so a negative keyword list built once and left alone degrades. Set a recurring reminder rather than relying on noticing a problem, by the time wasted spend is obvious in the headline numbers, it's been running for weeks.
Is Performance Max worth running alongside Search campaigns?
Only with brand exclusions in place and enough conversion data to feed it. Performance Max will absorb whatever budget it's given, and its reporting is deliberately limited, which makes it easy for it to claim conversions your Search campaigns would have produced more cheaply. Compare like for like before expanding it.
Why are my Google Ads getting clicks but no conversions?
Most often the landing page rather than the ads. Click your own ad on a phone: does the page answer the exact search, is a phone number visible without scrolling, does it load in under three seconds on mobile data, and is the form short? After that, check the search terms report, clicks from the wrong intent never convert regardless of the page.
Should I bid on my own brand name?
It's defensible when competitors are bidding on your name, and it shouldn't be your largest campaign by spend. Brand campaigns show flattering cost-per-conversion figures because many of those people would have found you anyway, which makes the account look healthier than it is. Measure it separately from the rest.
What's the fastest way to reduce wasted Google Ads spend?
Negative keywords from the search terms report, consistently. On one account we cut cost per enquiry by roughly a third within a fortnight through negatives alone, with no other change made. After that, check that Search and Display aren't sharing a campaign and that your geographic targeting excludes areas you don't service.