A common scene: Google Ads reports 40 conversions for the month, GA4 shows 28 key events from paid search, and the CRM says 31 real leads. Nobody is lying, and no one has necessarily made a mistake. These tools measure different things in different ways, and expecting identical totals is the first mistake to correct.
This article explains the main reasons numbers diverge and gives you a method to work out which differences are normal and which signal a real problem. The same logic applies when you compare GA4 with Meta Ads, Microsoft Ads or your store platform.
The tools count different things
Google Ads counts clicks on ads and the conversions credited to those ads. GA4 counts sessions and events on your site, from every source, and then assigns credit to channels. A click is not a session. One click can fail to load the page, or produce two sessions, and some visitors will never be recorded by GA4 at all.
Conversions and key events also differ by definition. A Google Ads conversion action can come from a GA4 import, a Google tag, a call from an ad, or an uploaded offline conversion. If your Google Ads account counts calls from ads or imported CRM leads, it will hold events that GA4 never sees.
Attribution models and credit
Attribution decides which touchpoint gets credit for a conversion. GA4's reporting attribution model is data-driven by default, which distributes credit across touchpoints using a machine learning model, with paid and organic last click options also available after Google retired the older rule-based models. Google Ads has its own attribution settings per conversion action, and by default it only credits Google Ads interactions.
That means a customer who clicked your ad, left, then returned through an organic search and converted may be credited fully to Google Ads in the ads interface, while GA4 gives some or all credit to organic search. Both are correct under their own rules.
- Google Ads credits only ad interactions it served
- GA4 compares credit across every channel it can see
- Lookback windows differ and can be configured separately
- Cross-channel credit adds up differently in each tool
Timing: conversion date versus click date
Google Ads typically reports conversions against the date of the ad click by default, while GA4 reports key events on the date the event happened. Take someone who clicks an ad on the 29th and enquires on the 3rd of next month. Google Ads files the conversion under the 29th, GA4 files it under the 3rd. For a single month, the totals will drift apart, especially for businesses with long consideration times.
Both tools also need time to process. Standard GA4 reports can take 24 to 48 hours to settle, and Google Ads conversions may be added later as modeled or delayed conversions come in. Comparing yesterday's numbers between the two is almost always misleading.
Consent, blockers and modeled data
When visitors decline analytics or ad storage in a consent banner, tags may be limited in what they can record. Where Consent Mode is implemented correctly, Google can use modeling to estimate some of the missing behavior, but eligibility depends on volume and configuration, and the modeled portion may appear differently in GA4 and Google Ads.
Ad blockers, browsers with strict tracking prevention, and slow pages that a visitor leaves before the tag loads also reduce GA4 counts. Adding tags through Tag Manager, loading them early and testing on mobile browsers helps, but it will not close the gap fully.
Thresholds, sampling and filtered views
GA4 can withhold rows of data when reports include demographic or interest information, or when Google signals is active, to protect user privacy. This is called data thresholding. Explorations can also apply sampling on very large queries. If a number looks too low in a report with extra dimensions, try a simpler report with fewer breakdowns, or check whether a thresholding icon is displayed near the report title.
Also review your internal traffic filters and any property or report-level segments. A comparison between a filtered GA4 report and an unfiltered ad account is never like for like.
A simple method to reconcile
Aim for explainable differences, not matching totals. Use a fixed process each month so the comparison stays honest.
- Pick a complete date range that ended at least three days ago.
- In Google Ads, segment conversions by conversion action and view them by the conversion time column, so dates line up with GA4.
- In GA4, filter key events by Session Google Ads campaign or by source and medium of google / cpc.
- Compare against a third source, such as the CRM or order system, to see which tool sits closer to reality.
- Note the typical gap. A stable gap is normal; a sudden change in the gap points to a tracking fault.
Once you know the usual gap, you can monitor it. A drop in GA4 key events with stable Google Ads conversions often means a tag issue, as covered in our GA4 setup checklist. The reverse can mean imported conversions are duplicating events.
If the differences are large and unexplained, a Google Ads and tracking review is worth doing before changing campaign strategy, since decisions made from broken measurement tend to compound. You can also reach us through the contact page to talk through your setup.
Key takeaways
- GA4 and Google Ads measure different things, so identical totals are not a reasonable goal.
- Attribution settings and date logic explain many of the differences between the two.
- Consent choices, blockers and thresholds can lower GA4 counts.
- Compare mature date ranges and use a third source such as your CRM as a reference.
- Monitor the gap over time: a sudden change usually means something broke.