The checks we run to prove GA4 and Google Ads conversion tracking work: a test enquiry, DebugView, Tag Assistant and a week of real leads.
To check your GA4 and conversion tracking are actually working, do one real test enquiry yourself and watch it arrive in GA4 DebugView, in Google Ads and in your inbox, then compare a week of real form submissions and calls against what each platform recorded. If the numbers line up within a small gap, your tracking is working; if one platform shows double, half or nothing, something is broken and your ad bidding is learning from bad data.
Not sure your tracking is right?
We set up and check GA4 and conversion tracking for Perth businesses.
This is the checklist we run, in the order we run it. If you would rather have someone do it for you, our conversion tracking and GA4 setup page explains how we work.
Google Ads and Meta both use your conversions to decide who to show your ads to and how much to bid. If the conversion signal is wrong, the platform optimises for the wrong thing, and it does so confidently.
That is why we treat tracking as the gate rather than an afterthought. Bidding is only ever as good as the signal it is fed, so on an inherited account the order is a full audit, then tracking repair, then structure. Dirty or unverified tracking stops a launch.
Write down every action on your site that matters to the business. For most Perth service businesses that is a short list: an enquiry form sent, a phone number tapped, a quote request, a booking, and for ecommerce a purchase. Next to each one, note where it happens, for example the contact page, the header call button or the checkout thank you page.
Then decide which one is the main goal. We set one primary valued conversion per goal and keep everything else secondary, so nothing double counts.
Start with the Realtime report in GA4. Open your site on your phone, browse a couple of pages and submit a test enquiry. Within a minute or so you should see yourself as an active user, with page views and your form or lead event listed.
Realtime tells you something is arriving. It does not tell you the details are right. For that, use DebugView, which sits under Admin in GA4 and shows each event from a device in debug mode, in order, with its parameters. The easiest way to put your browser into debug mode is to run Google Tag Assistant or Google Tag Manager preview mode on your site.
In DebugView, check three things for your test enquiry:
Be wary of the built-in form tracking in GA4's enhanced measurement. It can fire on forms you do not care about, such as a newsletter box or a search field, and it does not always know whether the submission went through. A dedicated event on the thank you page or on a confirmed success message is more reliable.
Google Tag Assistant connects to your site and shows every Google tag that loads and every event it sends, page by page. It is the quickest way to spot a site that has the GA4 tag hardcoded in the theme and again inside Google Tag Manager. That setup counts every page view twice and inflates everything downstream.
Walk through the full path a customer takes: land on a service page, open the contact page, submit the form, reach the thank you page. At each step, check which tags fired and which did not. A conversion tag that fires on the contact page instead of the thank you page will count every visit to the form as an enquiry.
If you run Meta ads, do the same check with the Test Events tab in Meta Events Manager. We check the pixel and conversion tracking before any Meta launch, so every result is measured properly.
In Google Ads, open Goals and then the conversions summary. Every conversion action has a status. "Active" is what you want. "Misconfigured" means it has stopped recording, and "Awaiting conversions" on an action you expect to be busy is a red flag worth chasing. "Needs attention" still records, but has a fix waiting.
Then check these settings on each action:
Phone calls deserve their own check. A tap on a phone number is not the same as a conversation. For phone driven trades we set valued call conversions of 60 seconds or longer, so a missed call or a wrong number is not counted as a lead.
This is the test that matters most, and the one most people skip. Pick a full week. Count the real enquiries that reached you: form emails in your inbox, entries in your CRM, booked jobs, answered calls. Then pull the same week from GA4 key events and from Google Ads conversions.
The numbers will not match exactly, and they should not be expected to. Ad blockers and privacy settings stop some visitors from being tracked at all. Google Ads also reports a conversion against the date of the ad click by default, not the date of the enquiry, so a lead on Monday from a click last Friday lands in a different day.
What you are looking for is the size and direction of the gap:
Our Impact Dents case study is reported as 1,100 deduplicated tracked enquiries at $9.53 each over twelve months, off one Google Ads Search campaign at around $30 a day. The word "deduplicated" is there on purpose. A cost per enquiry is only meaningful when each enquiry is counted once.
If customers book or pay on a separate platform, a click on your "Book now" button is not a booking. Plenty of businesses track the click out and call it a conversion, which counts everyone who looked at the booking page and left.
When bookings or payments happen on another site, such as Cliniko, Rezdy or Stripe, we track them there too, not just the click out. Check whether your booking or payment platform can send a confirmed event or redirect back to a thank you page on your own site, and test it with a real booking.
Recheck after anything changes on the site: a new form, a theme update, a page rebuild, a new booking system or a new developer. Those are the moments tags break without anyone noticing. A short real-enquiry test after each change takes ten minutes and saves weeks of bad data.
Outside of changes, run the week-of-leads comparison every few months. If an agency runs your ads, ask them to show it to you.
Tracking is included in our Google Ads management and Meta Ads management, which start at $800 including GST, billed every 4 weeks. Bought on its own, it is quoted to scope. Kobi Martin, one of our two founders, runs paid media and conversion tracking, and with read-only access we can check an account someone else built for double counting and orphaned actions before anything is changed. Details are on our conversion tracking page.
Submit a real test enquiry on your own site and watch it in GA4 Realtime and DebugView. Check the lead event fires once, only on a successful submission, and is marked as a key event. Then compare a week of real enquiries in your inbox or CRM with the key events GA4 recorded for that week.
The most common causes are a conversion counted twice, for example a Google Ads tag and an imported GA4 event both set as primary, a tag installed twice on the site, a lead action set to count every submission, or a tag firing on failed attempts and spam. Tag Assistant and the conversion settings usually show which one it is.
They measure differently. Google Ads reports a conversion against the date of the ad click by default, while GA4 reports the date the event happened, and the two use different attribution. Ad blockers and privacy settings also stop some visits being tracked. A small gap is normal; a gap of double or half means something is broken.
DebugView shows each event from a device in debug mode, in order and with its parameters, as it happens. It is used to confirm a tag fires once, at the right moment, with the right details. You can put your browser into debug mode by running Google Tag Assistant or Google Tag Manager preview mode on your site.
A tap on a phone number shows intent, but it is not a conversation. Many taps are missed calls or wrong numbers. For phone driven businesses we set valued call conversions of 60 seconds or longer as the main goal, and keep simple number taps as a secondary action so they do not drive the bidding.