How to tell if your Google Ads are actually working

I talk to small business owners about Google Ads a lot. The conversation usually starts the same way: “I’m spending money on Google Ads but I have no idea if it’s actually doing anything.”

That’s not a dumb question. Google’s dashboard throws dozens of metrics at you, most of them irrelevant to whether your phone is ringing more than it was before you started paying.

Here’s what I’ve learned from auditing Google Ads accounts for businesses across Melbourne: you only need to look at five numbers. Everything else is noise until those five are healthy.

What does “working” actually mean for Google Ads?

Google Ads is working when it generates leads or sales at a cost that leaves you with profit. That’s it.

Not impressions. Not clicks. Not “brand awareness.” Those are activity metrics, and they can all look fantastic while your campaign quietly burns money. I’ve seen accounts with thousands of clicks and a conversion rate of 0.3%. The dashboard looked busy. The phone was silent.

The question isn’t “are people seeing my ads?” It’s “are people seeing my ads, clicking on them, and then doing the thing I want them to do?” Calling you, filling out a form, booking an appointment, buying something.

I audited one account recently where the business owner was thrilled because impressions had more than quadrupled - from 12,000 to 50,000 per month. Great. But conversions had dropped from 15 per month to 8. The ads were being shown to more people, yes - just not the right people. Impressions without conversions is just expensive visibility.

A well-running campaign looks like this: you’re paying AUD $45 per qualified lead for a service worth $800 to you, and your search terms report is 90% relevant traffic. That’s the target. If your numbers look nothing like that, keep reading.

If you can’t answer that question from your dashboard, you probably don’t have conversion tracking set up properly. That’s the first thing to fix, and I’ll cover it in the audit checklist below.

The 5 numbers you need to check right now

These five run in order for a reason. The first two tell you whether your numbers can be trusted at all. The last three are the economics. Industry benchmarks appear throughout for context, and that’s all they are: they tell you what’s typical for your sector, not whether your campaign is healthy. That question gets answered by your own break-even maths, which is number five.

1. Tracking integrity - can you trust your numbers at all?

Before any other number means anything, your conversion tracking has to work. Go to Tools > Conversions. Are there active conversion actions? Do they show recent conversions? Google’s own documentation on conversion measurement explains how it works and links out to the setup instructions, and Stape has a detailed guide (updated April 2026) covering both client-side and server-side tracking methods.

The quickest test: fill out your own contact form, or call your own tracking number. If a conversion doesn’t appear in your account within 24 hours, something is broken, and every other number in this article is fiction until you fix it.

2. Qualified leads - is the phone ringing with real enquiries?

A “conversion” in the dashboard is a recorded action. A qualified lead is someone you’d actually quote. They’re not the same thing, and Google will cheerfully count spam form submissions, wrong numbers, and job seekers as conversions. Cross-check the dashboard against your inbox, call log, or CRM for the same period, and count the enquiries you’d genuinely want more of. That count, not the platform’s, is what the rest of the maths runs on.

3. Conversion rate - are visitors actually doing something?

This is where benchmarks start earning their keep. WordStream analysed 13,474 US search advertising campaigns across 23 industries between April 2025 and March 2026 and found a Google Ads average conversion rate of 8.18%. If yours is below 2%, look hard at your landing page and your offer. Paid clicks just make a leaky page more expensive, so it’s worth checking where enquiries are actually escaping - I run through the seven most common spots in before you pay for SEO: 7 enquiry leaks to check first.

Auto repair and service businesses convert at 15.51% — near the top of the table (only pets-and-animals businesses rank higher). Real estate sits down at 3.70%, and finance and insurance is the lowest of all at 2.64%. Know your industry benchmark before you panic about yours.

4. Cost per qualified lead - what does each real enquiry cost you?

Divide your ad spend by the qualified-lead count from number two, not by the dashboard’s conversion total. For context, WordStream’s 2026 benchmarks put the average cost per lead at $66.69 USD, with a huge spread by industry: arts and entertainment at the bottom on $26.84, shopping, collectibles and gifts at $49.40, home improvement at $90.92, and attorneys and legal services at the top on $131.63 (WordStream, 2026).

Those are US dollars from US campaigns. A note on Australian figures: an earlier version of this article carried a per-industry AUD cost-per-lead chart credited to an industry report. When I traced the chain for this update, the agency roundup it came from actually attributes those numbers to a blend of US benchmarks and “agency-aggregated data”, not a published Australian study, so the chart’s gone. I haven’t found reliable public AUD lead costs by industry. If you see them in an agency roundup, treat them as estimates.

If your cost per qualified lead is more than double your industry average, that’s worth investigating. If it’s below average, you might have room to increase your budget and get more leads at a price that works. Either way, the benchmark is context. The comparison that actually decides things is the next number.

5. Return against break-even - does the maths actually work?

This is the number no dashboard shows you, and it’s the one that decides whether the ads are “working.” Work backwards from what a customer is worth. If a customer is worth $500 to you and a qualified lead costs $70, you can afford about seven leads for every customer you win. Close better than one in seven and the ads are making you money. Close worse and they’re not, no matter how busy the dashboard looks.

Google’s own estimate is that businesses make $2 in profit for every $1 they spend on Google Ads. That figure came from their Economic Impact methodology (developed by Google’s first chief economist, Hal Varian, back in 2009), so take it with a grain of salt - it’s Google measuring the value of Google. Google has since taken that page down, so the link goes to an archived copy. One Australian agency roundup puts average return on ad spend at 5.2x (3P Digital, which describes the figure as aggregated from Skai and WordStream data). Treat it as a rough industry aggregate, not a target. Plenty of businesses are below it, and plenty are above it.

Neither average pays your invoices. The question is whether your return clears your break-even, on numbers you’ve cross-checked yourself.

Two supporting diagnostics: CTR and CPC

Click-through rate and cost per click usually get top billing in articles like this. They shouldn’t. Neither tells you whether the ads are working. They tell you where to look when one of the five numbers above is off.

Click-through rate (CTR) tells you whether your ad copy and targeting match what people are searching for. The 2026 Google Ads average is 6.64% (WordStream, 2026); restaurants and food businesses average 6.83%, attorneys 5.87%, so compare within your industry, not across it. A low CTR usually means one of two things: your keywords don’t match your ads, or your ad copy isn’t convincing enough to get the click. Sometimes both. I’ve seen accounts where the ads were actually decent, but they were showing up for searches that had nothing to do with the business. That’s a targeting problem, not a copy problem, and the fix is different.

Cost per click (CPC) is what you pay for each visitor. The 2026 Google Ads average is $5.42 USD (WordStream, 2026). The Australian-dollar figures you’ll see around (an average near AUD $3.60, ranging from about $1.20 for e-commerce to $12.80 for legal and financial services) are an agency compilation built from those same US benchmarks, so treat them as estimates (3P Digital). A high CPC isn’t automatically bad. A dentist paying $8.00 per click but converting 10.67% of those clicks into patients is doing fine. A retailer paying $8.00 with a 1% conversion rate is not. CPC only means something next to conversion rate and lead value.

Where does Quality Score fit? (Not where you might think)

Quality Score is Google’s 1-10, keyword-level rating of how relevant your ads and landing page are, built from three components: expected CTR, ad relevance, and landing page experience. For years, articles like this one treated it as a headline health metric. An earlier version of this article did too. Google’s own guidance is blunt about what it actually is: “Quality Score is not a key performance indicator and should not be optimized or aggregated with the rest of your data”, and it “is not an input in the ad auction”. It’s a diagnostic tool.

So use it the way Google says to: for troubleshooting. The ad quality Google calculates at auction time does affect what you pay and where you show, and the three QS components point you at what to improve. If landing page experience reads “below average,” the page your ad points to probably doesn’t deliver what the ad promised, and Core Web Vitals (page speed and visual stability) feed into that experience too. Fix the component, not the number.

How to read your Google Ads dashboard without losing your mind

Google’s interface changes constantly, and it shows you way too much data by default. Here’s how I navigate it when I’m auditing an account.

Start in Campaigns, not Overview. The Overview tab is Google’s curated highlight reel. It tells you what Google wants you to see, not necessarily what you need to see. Go to Campaigns instead.

Add the columns that matter. By default, you’ll see impressions and clicks. You need to add: CTR, Avg. CPC, Conversions, Conv. rate, Cost/conv., and Quality Score. Right-click on any column header and select “Modify columns” to add them.

Filter by date range. Look at the last 30 days first. If you’ve been running ads for a while, compare to the previous 30-day period. Are things getting better or worse?

Check the Search Terms report. This is the most revealing report in the entire dashboard, and it’s the one most business owners have never looked at. It shows you the actual words people typed before they saw your ad. Go to Keywords, then Search Terms.

You will almost certainly find searches in there that have nothing to do with your business. Every click on an irrelevant search is money gone. I once audited an account for a Melbourne landscaper who was spending $1,200 per month. About $400 of that was going to clicks from people searching “landscape painting classes” and “landscape photography tips.” Different kind of landscape. After adding negative keywords and tightening match types, their cost per lead dropped from $95 to $42.

The fix is adding negative keywords - telling Google “don’t show my ad when someone searches for this.” It takes ten minutes and can save you hundreds of dollars a month.

Brad Geddes, one of the most respected PPC experts in the industry, puts it this way in his 2026 Search Engine Land article: not adding your top search terms as keywords and applying old negative-keyword lists without reviewing what’s in them are two of the top 10 mistakes he sees across Google Ads accounts.

The 5 most common ways small businesses waste their ad budget

I’ve borrowed from Brad Geddes’ list (Search Engine Land, January 2026) and added what I see specifically in Australian SMB accounts.

1. No conversion tracking (or broken tracking)

If you don’t know how many people called you, filled out your form, or booked an appointment after clicking your ad, you can’t tell if anything is working. You’re flying blind. It’s number one on the scorecard above for a reason: nothing else in this list matters until it’s fixed.

2. Ignoring the Search Terms report

I mentioned this above because it’s the single biggest source of waste I see. I regularly find 20-30% of ad spend going to completely irrelevant searches. One Melbourne tradie I audited had a monthly spend of $2,100 - and $630 of it was going to clicks from people searching for DIY tutorials. Those people were never going to hire him. We cut that waste in the first week just by adding negative keywords.

3. Blindly accepting Google’s recommendations

Google’s Optimisation Score and auto-applied recommendations are designed to get you to spend more. Geddes says it plainly:

“Google’s recommendations are sometimes in your best interest. They are always in Google’s best interest.”

• Brad Geddes, Search Engine Land, January 2026

Some recommendations are fine. Some will broaden your targeting to irrelevant audiences or switch you to ad formats that don’t suit your business. Check each one individually. Don’t hit “Apply all.”

4. Using broad match keywords with automated bidding and no guardrails

Broad match can work, but only if you pair it with a target CPA or target ROAS bidding strategy, and you’re watching the Search Terms report closely. Without those guardrails, broad match will happily show your ad for searches that have nothing to do with what you sell.

5. Ignoring what Quality Score is trying to tell you

Quality Score is a diagnostic, not a target (see above), but a keyword sitting at 3 with a “below average” component is Google telling you exactly where the relevance problem is. Make sure your ad copy matches your keywords, and that your landing page actually delivers on what the ad promises. It’s not magic. It’s relevance.

What your ads cost in Australia (and whether that’s normal)

Australian internet advertising spend reached $17.2 billion in FY25, up 10.6% year-on-year, and search holds the largest share of that market at 44% (IAB Australia’s Internet Advertising Revenue Report, prepared by PwC Australia).

So if your competitors are advertising anywhere online, search is probably where. The question is whether they’re doing it well. The same buyer-protection logic applies to organic search, by the way - if you’re weighing up an SEO retainer alongside your ad spend, I’ve written a guide on what SEO costs in Australia and how to read the quote before you sign anything.

If you’re running ads to a Melbourne audience, your local SEO foundation determines whether the same searches also surface you organically — saving you from paying twice (once for the ad click, then nothing for the equivalent organic visit). Get the Map Pack right and your paid ads stretch further.

Whether the spend comes back is the break-even question in number five of the scorecard above, and you can only answer it with tracking that works and a qualified-lead count you trust. If you’re not sure about the technical health of your site more broadly, I wrote a separate guide on Core Web Vitals and what they mean for your business.

A 15-minute self-audit checklist

You can run through this in one sitting. You’ll need access to your Google Ads account and about 15 minutes.

1. Verify your conversion tracking. Go to Tools > Conversions. Are there active conversion actions? Do they show recent conversions? If this page is empty or the last conversion was months ago, stop here and fix this first.
2. Pull up the Search Terms report. Go to Keywords > Search Terms. Sort by cost. Are the top 10 most expensive search terms relevant to your business? Add irrelevant terms as negative keywords.
3. Count your qualified leads and work out what each one costs. Cross-check the dashboard’s conversions against your inbox or call log for the last 30 days, then divide your spend by the real number. Compare that to what a customer is worth to you. The industry benchmarks in this article are context if you’re curious how you sit against your sector; they’re not the verdict.
4. Scan Quality Score for troubleshooting clues. Add the Quality Score column to your Campaigns view. It’s a diagnostic, not a score to chase: for keywords sitting at 3 or 4, check which component (expected CTR, ad relevance, landing page experience) reads “below average” and fix that.
5. Review Google’s recommendations with scepticism. Go to Recommendations. Read each one. Ignore the Optimisation Score number - it’s designed to get you to spend more. Only apply recommendations that make sense for your specific business.

If you get through this checklist and everything looks healthy, your ads are probably working fine. You might still want to optimise, but at least you know the foundation is solid.

If you find problems at multiple steps, it might be worth getting a professional audit. Most of the issues I find in SMB accounts can be fixed in a few hours. The hardest part is usually the first one - getting conversion tracking working properly. Everything after that is tuning.

One more thing: check which attribution model your reporting uses. If it’s last-click, you’re only crediting the very last click before a sale, and that can make Google Ads look worse than it actually is. If a customer saw your ad, clicked it, then came back three days later through Google organic and bought - last-click gives Google Ads zero credit. Something to keep in mind when you’re reading your numbers.

If your numbers are off, here’s what to fix first

If you ran the audit and found problems, don’t try to fix everything at once. Start with whichever number is doing the most damage to your break-even maths and work from there.

• Tracking you can’t trust: Fix this before touching anything else. Re-run the self-test from number one on the scorecard; every other number is fiction until it passes.
• Low conversion rate (below 2%): The problem is usually the landing page or the offer. Make sure the page delivers exactly what the ad promised, loads fast on mobile, and has a clear call to action above the fold — if the copy on the page sounds like anyone could have written it, here’s how to fix that. Check your site speed and Core Web Vitals while you’re at it.
• High cost per qualified lead (well above what your break-even allows): This is usually a combination of the other issues on this list. Fix conversion rate and ad relevance first, and cost per lead will come down.
• Low CTR (well below your industry’s average): Your ads aren’t matching what people are searching for. Rewrite your ad copy to more closely match your keywords, or tighten your keyword targeting. Check that your ad groups aren’t mixing unrelated keywords together.
• High CPC (well above your industry’s average): Check your ad relevance and landing page experience components first - the ad quality Google calculates at auction time affects what you pay. Add negative keywords to stop paying for irrelevant clicks. If quality looks fine, you might just be in a competitive market, and the fix is improving your conversion rate instead.

Full disclosure: I offer Google Ads audits as a service. The framework above works regardless of who implements it - if you’re comfortable in the dashboard, you can absolutely do this yourself. I wrote this article so you’d have the benchmarks and the process to check your own campaigns without needing to hire anyone.

Frequently asked questions

If you’d like me to run through a Google Ads audit for your business, get in touch. You can read more about how I work or go straight to booking a call. I work with businesses across Melbourne and I’m always happy to have a look.

For more on how Google Ads connects to the rest of your search visibility: Core Web Vitals for small businesses, local SEO for Melbourne businesses, and GEO and AI search optimisation cover the same territory from different angles.

All benchmarks in this article were verified against their original sources in July 2026 (WordStream figures updated to the 2026 edition, with two sector figures corrected against it, and the Australian market figures now cited to IAB Australia directly). Also updated July 2026: the scorecard was restructured. An earlier version treated Quality Score as one of the five headline numbers, and per Google’s own guidance it’s a diagnostic tool, so the article now treats it that way. Industry benchmarks are typically updated annually - check the linked sources for the latest figures.