GA4 and Google Ads: why conversions don’t match up
The discrepancy between conversions in Google Ads and those in GA4 is normal and comes down to three different things: a different attribution model, a different reporting window and a different moment at which each platform records the sale. Google Ads attributes the purchase to the day of the click, GA4 records it on the day it actually happened. That is why, week by week, the two numbers rarely match. A difference of up to about 20% is usually explainable and needs no intervention. A larger difference almost always means a tracking defect, not a statistical one: duplicated tags, missing consent, a break during payment or different definitions of the same event. Then the number you are making decisions on is wrong.
My name is Hristiyan Panov — a performance marketing consultant with over 5 years’ experience and over €2 000 000 Managed advertising budgets for over 90 companies on Google and Meta. Below are the reasons for the discrepancy, which figure you should use when making decisions, and how to distinguish between normal and abnormal results.
Why do the two platforms give different figures?
- Different attributions. Google Ads attributes the conversion of the day to the click. GA4 records it on the day of the purchase. If someone clicks on Monday and makes a purchase on Thursday, the two systems record the transaction on different days.
- Different scope. Google Ads only counts conversions that follow an interaction with an advert. GA4 counts them all, regardless of the channel.
- Different approaches to handling missing data. Google Ads estimates conversions where consent is not given. GA4 reports the actual data it has received.
- Various windows. The standard time frame in Google Ads is 30 days. If this does not match the time frame in GA4, you are comparing different things.
What is a normal difference?
A rule of thumb from audits: up to 20% is understandable for the reasons mentioned above. Between 20 and 40%, I’m already looking for a specific cause. Above 40%, I assume that something is broken until proven otherwise.
The percentage only makes sense if it is measured correctly. Compare the same period, the same time zone and the same event, and in GA4 filter only the traffic from Google Ads. Do not measure by a single day — with a 30-day window an individual day is noise. Look at a whole month and watch whether the difference is stable.
- A stable difference in one direction — most often attribution and modelling. It is explainable.
- A difference that jumps from week to week — more likely a tracking defect.
- A matching count but diverging revenue — a problem with the value or the currency, not with attribution.
Important note: compare with actual orders in the shop, not the two advertising platforms pitted against one another. The ordering system is the only source that does not model or attribute.
What are the five most common reasons for a significant discrepancy?
- Duplicate entries. The same event is sent from several sources — the tag in the theme, a plugin and Tag Manager — all at the same time.
- Missing or incorrect agreement. Without a proper consent signal, GA4 loses data, whilst Google Ads models it — the two systems differ in design.
- Counting the incorrect event. A conversion is defined as an event that is not an order — for example, a visit to a thank-you page that reloads.
- Order value missing. The numbers match, but the revenue does not. Value-based optimisation becomes impossible.
- Payment interruption. Payment via a third-party provider, which redirects the user without the session parameters — the conversion is attributed to the wrong source.
Which figure should you use when making decisions?
It depends on the decision.
- For campaign management — Google Ads. The algorithm optimises based on what it sees, so decisions regarding budgets and bidding are made in line with its logic.
- For distribution between channels — GA4. It gives you the full picture, not just the paid part.
- For profit — the ordering system. That’s the only place where you’ll find actual figures, returned goods and the actual profit margin.
The most common costly mistake is the opposite: profit-driven decisions based on the advertising platform’s figures.
How to check in half an hour
- Set the same date and time zone in both systems.
- In GA4, filter for traffic from Google Ads only so that you can compare like with like.
- Compare both figures with the number of actual orders for the same period.
- Check whether there is more than one source sending the same event.
- Please check that the order value is being transferred and that the currency is the same.
- Place a test order and check that it appears exactly once throughout.
When does a discrepancy mean a financial loss?
When the error is systematic. If the account consistently sees fewer conversions than there actually are, it will automatically limit spending on profitable campaigns. If it sees more, it will increase spending on unprofitable ones.
The damage accumulates because automated bidding learns from the same signal. One day of wrong data is compensated for. A month of wrong data rearranges the whole budget, and after the fix a period of re-learning follows — the loss is double.
The most dangerous case is when the gap is uneven: only for some of the devices, browsers or payment methods. Then the platform does not underestimate everything equally, but shifts the budget towards the segments that are simply measured better.
- Compare at campaign level, not only overall — the total can match while individual campaigns err in opposite directions.
- Check whether campaigns stopped as losing have orders in the order system for the same period.
In both cases the problem is not solved with budget. It is solved by fixing the tracking — and that is the first thing I do in every account audit.
FAQ — Frequently Asked Questions about GA4 and Google Ads
Why does GA4 show fewer conversions than Google Ads?
This is usually because Google Ads adds modelled conversions where consent is not given and attributes them by click date, whilst GA4 only reports actual events by purchase date.
Can the two be brought into complete alignment?
No. They measure different things in different ways. The aim is for the difference to be explainable and stable, not zero.
What should I include in the report for the client?
Actual orders from the shop serve as the basis, whilst advertising figures are used to explain the channel’s contribution. This ensures the report remains accurate even when the figures from the different platforms differ.
Does server-side tracking help?
Yes, because it restores some of the events that the browser blocks. It does not eliminate the discrepancy in attribution, but it reduces data loss.
How can I tell if there are duplicates?
Place a test order and check how many times it appears. If it appears more than once, there are several sources of the event.
Conclusion
The discrepancy between GA4 and Google Ads is not a fault in itself. It is a fault when you are unaware of it why. differ, and by how much — because then every budgetary decision is based on an assumption.
If the discrepancy in your case is over 20% and no one can explain it, start by arranging an independent review of the tracking.
Request a quote for a digital audit →
Related articles: Consent Mode v2 · server-side tracking · Google Shopping
