For over a decade, Universal Analytics was the default answer to “how do we track our website?” It was familiar, it was free, and nearly every marketer who touched digital analytics learned their trade inside its interface of sessions, pageviews, and bounce rates. Then, in March 2022, Google announced that Universal Analytics would be sunset, with standard properties ceasing to process new data on July 1, 2023, and Universal Analytics 360 properties following on October 1, 2023 (later extended to July 1, 2024). Historical data remained accessible for a limited window afterward before being permanently deleted.
That transition forced millions of businesses to adopt Google Analytics 4, whether they were ready or not. Even years later, many marketers who learned analytics on Universal Analytics still find themselves confused by GA4’s very different structure, terminology, and reporting logic. This guide breaks down exactly what changed, why Google made these changes, and what it means for how you measure your website and app performance today.
Why Google Replaced Universal Analytics in the First Place
Understanding what changed is easier once you understand why Google made the switch. Google pointed to three main forces driving the shift: rising privacy regulations, structural limitations in how Universal Analytics measured user behavior, and the need to prepare for a future where cross-device and cross-platform measurement, not just single-website tracking, would be the norm.
Universal Analytics was built in an era when most customer journeys happened entirely on desktop, within a single website, using cookies that reliably persisted for long periods. By the early 2020s, that world had changed dramatically. Customers increasingly moved fluidly between mobile apps and websites, browsers were cracking down on third-party cookies, and privacy regulations like GDPR and CCPA demanded far more careful handling of user data. Universal Analytics, at its architectural core, simply wasn’t built to handle that reality gracefully. GA4 was designed from the ground up with these constraints in mind, even if the transition proved painful for many users along the way.
The Core Difference: Sessions vs Events
The single biggest conceptual shift between the two platforms is the data model itself. Universal Analytics was fundamentally session-based and hit-based. It grouped user interactions into sessions, meaning a defined period of activity, and classified each interaction as a specific “hit type,” such as a pageview hit, an event hit, or a transaction hit. This structure was rigid: pageviews were pageviews, events were events, and the two lived in somewhat separate reporting worlds.
GA4 abandons this structure entirely in favor of an event-based model. In GA4, everything is an event. A pageview is an event. A scroll is an event. A video play, a file download, a button click, a purchase, a form submission: all of these are simply events with different names and different parameters attached to them. This is a much more flexible structure, since it means you’re not limited to Google’s predefined hit types; you can create and customize events to track virtually any interaction that matters to your business.
The practical implication is significant. In Universal Analytics, comparing “how many people viewed a page” versus “how many people submitted a form” required navigating two entirely different report categories. In GA4, both are simply events, viewable and comparable within the same events report, filtered however you like.
What Happened to “Sessions” in GA4
Sessions still exist in GA4, but they’re defined differently. In Universal Analytics, a session was a period of time, typically ending after 30 minutes of inactivity or at midnight, during which a user’s interactions were grouped together. In GA4, a session is generated by a specific “session_start” event, and the platform calculates session duration based on the time between the first and last event within that session, rather than a fixed timeout window in the same rigid way.
This might sound like a subtle distinction, but it can cause session counts to differ noticeably between the two platforms when comparing historical data, which was a major source of confusion (and no shortage of frustrated support tickets) during the transition period.
Automatically Collected and Enhanced Measurement Events
One meaningful improvement in GA4 is what Google calls Enhanced Measurement. When you set up a GA4 web data stream, the platform automatically begins tracking a set of common interactions without requiring any additional configuration: page views, scrolls, outbound clicks, site search usage, video engagement, and file downloads are all captured by default. In Universal Analytics, tracking most of these required custom event configuration through Google Tag Manager, meaning many businesses simply never bothered to set them up and were left with pageview data alone.
This is a genuine convenience win for GA4, particularly for smaller businesses or solo website owners who don’t have a dedicated analytics implementation team. That said, GA4’s default data collection is still comparatively limited for anything beyond these standard interactions, meaning meaningful custom event tracking (like tracking specific product interactions in an ecommerce store) still requires deliberate setup, typically through Google Tag Manager.
Conversions Replace Goals
Universal Analytics used “Goals” to define what counted as a conversion, and it limited most properties to twenty goals per view. GA4 replaces this concept with “Conversions,” which are simply events you’ve marked as important. Standard GA4 properties can mark up to thirty events as conversions, while GA4 360 (the enterprise tier) allows up to fifty. Because everything in GA4 is already an event, marking something as a conversion is often as simple as flagging an existing event, rather than building a separate goal configuration from scratch as you had to in Universal Analytics.
A New Interface and a New Way of Exploring Data
The most immediately noticeable difference for longtime Universal Analytics users is simply how different GA4 looks. Universal Analytics organized data into a fairly rigid hierarchy of standard reports: Audience, Acquisition, Behavior, Conversions, each with numerous sub-reports. GA4 consolidates and restructures much of this, emphasizing a “Reports snapshot” home view with automated insights generated by Google’s machine learning, alongside a more flexible “Explore” section.
Explorations are arguably one of GA4’s most powerful additions. Rather than being confined to a handful of predefined report templates, Explorations let you build custom analyses using templates like Funnel Exploration, which lets you visualize and analyze multi-step user journeys and identify exactly where users drop off, and Path Exploration, which uses tree-graph visualizations to show the different paths users take through your site or app. These tools simply didn’t exist in Universal Analytics, and for teams willing to invest the time to learn them, they offer considerably deeper behavioral insight than anything the older platform provided.
The tradeoff is a steeper learning curve. Many users, particularly those who relied heavily on Universal Analytics’s simpler, pre-built standard reports, found GA4’s interface and terminology genuinely disorienting at first, and some marketers still describe missing features or preferring the old platform’s straightforwardness even years after the switch.
Cross-Platform Measurement: Web and App Together
One of GA4’s most substantial architectural improvements is native support for measuring both websites and mobile apps within a single property. Universal Analytics required separate properties for web and app tracking (the app side historically running through Firebase Analytics), making it genuinely difficult to build a unified view of a customer who might browse on their phone’s app one day and complete a purchase on desktop the next.
GA4 was built around Google’s underlying Firebase-based measurement protocol from the start, allowing businesses with both a website and a mobile app to see the combined customer journey across both platforms within one unified data set. For any business operating across both an app and a website, this alone is often reason enough to fully embrace GA4’s data model rather than trying to recreate Universal Analytics workflows within it.
Attribution Model Changes
Universal Analytics offered a range of attribution models that users could apply retroactively within reports, including first-click, linear, time-decay, and position-based models, giving flexibility in how credit for a conversion was distributed across the touchpoints in a customer’s journey. GA4 has moved away from this multi-model flexibility in standard reporting, deprecating several of the older attribution models and defaulting most reporting to a data-driven attribution model, which uses machine learning to distribute credit based on actual observed patterns in your own conversion data, rather than a fixed, rule-based formula.
For marketers accustomed to manually comparing first-click versus last-click attribution in Universal Analytics, this shift requires some adjustment, though data-driven attribution is generally considered more accurate for businesses with sufficient conversion volume to train the model effectively.
What This Means for Your Business Today
If you’re still working from old Universal Analytics habits, the most important mental shift is to stop thinking in terms of pageviews and sessions as your primary unit of measurement, and start thinking in terms of events and the specific business actions you actually care about. Take time to configure custom conversion events that map to what genuinely matters for your business, whether that’s a completed purchase, a submitted lead form, or a specific engagement milestone, rather than relying purely on the automatically collected defaults.
It’s also worth investing time in GA4’s Explorations, particularly Funnel and Path explorations, since these represent genuine analytical capability that simply wasn’t available in the old platform. Many businesses that migrated reluctantly during the 2023 sunset never went back to properly explore these tools, and are consequently missing out on insights their data has been capable of providing for years.
Finally, remember that any historical Universal Analytics data you didn’t export before it was deleted is gone permanently. If you’re building reports or dashboards that reference year-over-year comparisons spanning the transition period, you’ll need to work with whatever exports you preserved at the time, since GA4 cannot retroactively recreate Universal Analytics’s historical session-based data using its own event-based model.
The transition from Universal Analytics to GA4 was, without question, one of the more disruptive changes in the history of web analytics tooling. But the underlying direction, toward event-based flexibility, cross-platform measurement, and privacy-conscious data collection, reflects where digital measurement was always heading. Understanding these differences isn’t just about nostalgia for an old interface; it’s about making sure you’re actually using the tool you have access to, rather than trying to force it to behave like the one it replaced.
