Three numbers for the same month.
This is what we see in almost every company before we start:
None of these tools is lying. They answer different questions. The trouble starts when nobody can say which part of the gap is expected and which part is a leak. Then every meeting begins with the same question: which one is right?
Where the gaps come from.
Attribution: every ad platform gives itself credit for a sale that happened after one of its ads. If a customer clicked a Google ad and saw a Meta ad, both report the sale. GA4 shares the credit across channels, so it shows fewer per channel.
Timing: Google Ads reports a conversion on the day of the ad click, GA4 and the shop on the day of the purchase. A click on the 30th and an order on the 2nd land in different months.
Consent and browsers: visitors who decline cookies, use ad blockers or browse in Safari are partly invisible to GA4. Google Ads fills some of that gap with modeled conversions, the shop sees every order. Three tools, three different blind spots.
Implementation: the purchase tag fires twice when the thank-you page reloads, fires before the payment fails, or has no transaction ID to remove duplicates. Revenue is sent gross in one tool and net in another. These are not differences in method. These are leaks.
Returns: the shop subtracts cancellations and returns. The ad platforms never hear about them unless you tell them.
Which gaps are normal, and which cost you money.
Why this matters: smart bidding optimizes on the number it sees. If Google Ads counts duplicates, it pays for customers it already had. If it misses a third of the sales in one market, it moves budget away from the market that actually sells. Nobody notices, because the dashboard still looks busy.
See in 30 seconds whether your purchase setup leaks.
How to get to one number everyone trusts.
The goal is not identical numbers in every tool. That will never happen, and anyone who promises it is guessing. The goal is a gap you can explain, that stays stable, and one agreed source of truth for every meeting.
Behind most stubborn gaps there is a missing measurement plan: nobody ever wrote down what a conversion is, where it is counted and who owns it.
From a client who brought us back.
“If you need analytics you can actually trust, Tom from Web Wombats is the person to call. I worked with him as an external partner at Medela and brought him back at Ypsomed because he delivers every time.”
Margarita Colombo, Associate Director Digital Marketing at Ypsomed, before that Global Digital Marketing Manager at Medela
Questions about conversion discrepancies.
Why does Google Ads show more conversions than GA4?
Google Ads credits itself for every conversion after an ad click within its conversion window, reports on the click date and adds modeled conversions for visitors without consent. GA4 shares credit across channels and counts on the session date. A difference of this kind is expected. Many more conversions in Google Ads than orders in your shop is not, and usually points to duplicates.
Why does GA4 show less revenue than our shop?
GA4 only sees visitors it is allowed and able to track. Declined cookies, ad blockers and browser limits remove part of the sessions, so GA4 revenue is usually lower than the shop. If the gap is large or changes from month to month, check that every order sends a purchase event with a transaction ID and the right value.
Which number should we report to management?
Revenue and orders from the shop or ERP, because that is money that actually arrived. Platform numbers are for steering campaigns inside each platform. Mixing the two in one report is where most arguments start.
Can server-side tracking close the gap?
Partly. Server-side tracking recovers sales lost to ad blockers and browser limits for visitors who consented, and it makes duplicates easier to control. It does not make visitors who declined cookies visible, and it should not.
How long does it take to find the cause?
The obvious leaks show up in a reconciliation of one week of orders. Finding the cause per market and fixing it properly is the first part of a tracking audit.
Keep reading
google ads 268 conversions
ga4 191 conversions
orders 240 Why every platform shows you different numbers
They will never match, and none of them is lying. What each number counts, and which one to use for what.
browser → metrics.yourdomain.comsGTM → GA4 · Ads · Meta CAPI
consent forwarded + enforced Should you run tracking on your own domain?
The honest decision guide: what first-party tagging changes for your ad performance, what it costs, and when to skip it.