Google Analytics 4 (GA4)
Google Analytics 4 is Google's analytics platform, built around individual events rather than the sessions-and-pageviews model that preceded it, letting you track any action a visitor takes as its own record rather than a page load.
Why it matters
It is where you see what happened after the click across the whole site, not just the single conversion action the ad platform recorded. It shows the path visitors took, the pages that lost them, and behaviour on channels you are not even paying for.
What people get wrong
Expecting GA4 conversion counts to match Google Ads exactly. They will not, and neither report is broken - the two use different attribution models, different windows, and sometimes different definitions of what counts as a conversion. Reconciling them to the decimal is not a goal worth pursuing; understanding the gap is.
How it actually works
GA4 records every interaction as an event with parameters attached - a page view, a scroll, a click, a form submission, a purchase - rather than pre-defining a rigid hierarchy of sessions and pageviews. This is a deliberate departure from Universal Analytics, and it means almost any user action can become a trackable event if someone configures it to fire.
Events reach GA4 either automatically (page views, scrolls, outbound clicks are collected by default) or through explicit tagging, usually via Google Tag Manager, which listens for a specific trigger on the page and pushes a named event with whatever details matter - a form ID, a button label, an order value.
From there GA4 assembles those events into a picture of a session using its own modeled attribution, which is not identical to how Google Ads attributes the same conversion internally, and it applies data-driven or last-click models depending on report and configuration, which is a second reason the numbers diverge from the ads platform.
What goes wrong in real accounts
The single most common problem is duplicate or missing conversion events - a form that fires an event on every page load instead of on submission, inflating conversions, or a checkout confirmation page that only fires the event when loaded directly and never when the visitor arrives through a normal flow, silently dropping conversions. Both look fine on a dashboard that only shows totals.
A second problem is treating GA4's default event set as sufficient. Out of the box it captures generic engagement, not business outcomes - it does not know that a click was a phone number, a request for an estimate, or a completed booking unless someone marks it as a conversion. An account that never configures this has a GA4 property full of activity and no way to say which of it mattered.
A third is data sampling and threshold suppression in reporting views, particularly around demographics and smaller segments, which can make a channel look like it produced nothing when it actually produced too few events to display under GA4's privacy thresholds. This gets misread as a channel failing rather than a reporting limitation.
How it relates to the other terms
GA4 usually receives its events through Google Tag Manager, which is the delivery mechanism for almost everything that is not automatically collected, so a broken tag in the container breaks the GA4 data downstream without GA4 itself showing an error.
It sits alongside, not inside, platform-level conversion tracking - Google Ads has its own conversion actions and its own attribution window, and GA4 is a separate lens on the same visitors, useful precisely because it is not trying to optimize bids the way the ads platform is.
And when someone asks why the conversion rate reported in GA4 does not match the one in the ads account, the answer is almost always a combination of attribution model and event definition, not a bug in either system.
Frequently asked
Why don't GA4 conversions match Google Ads conversions?
They use different attribution models and often different definitions of what counts as a conversion, plus different windows for how long after a click a conversion is credited. This is expected behaviour, not a sign that either system is broken. Use each for what it is good at - GA4 for the fuller behavioural picture, Ads for what it is actually optimizing toward.
Do I need Google Tag Manager to use GA4 properly?
Not strictly, but most businesses do because GA4's automatic collection only covers generic engagement events, not business-specific actions like a quote request or a completed booking. Tag Manager is the usual way to fire those custom events reliably without editing site code every time something changes.
Why does a channel show almost no conversions in GA4 when I know it's producing them?
Check whether the conversion event is actually configured and firing on the right page, since a misconfigured event will simply not record anything with no visible error. If the event is confirmed working, look at data thresholds and sampling in smaller segments, which can suppress low-volume rows in reporting without removing the underlying data.