Google Analytics 4 New Session When Campaign Source Changes

If you're used to Universal Analytics, you’ll remember one particular quirk: a new session would start whenever a user's campaign source changed mid-visit. So, does Google Analytics 4 handle things the same way? The short answer is no, it doesn’t. This article breaks down how GA4 sessions work, why this change was made, and what it really means for your marketing analytics.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

First, A Quick Look Back: Sessions in Universal Analytics

To appreciate the change in GA4, it’s helpful to remember how things worked in Universal Analytics (UA). In UA, a session was a collection of hits (like pageviews or events) from a single user. A session would end for one of two main reasons:

1. Time-based expiration: This is the one we’re all familiar with. If a user was inactive on your site for a set period (usually 30 minutes), the session would time out.
2. Campaign override: This is the key difference. If a user arrived on your site from one campaign (say, a Google CPC ad), browsed a few pages, and then clicked a link in your newsletter to come back - all within 30 minutes - UA would close the first session and start a brand-new one attributed to the "email" source.

Here’s a practical example of the old UA override logic:

• 10:00 AM: A user clicks a Google Ad (Source/Medium = google / cpc) and lands on your homepage. Session 1 starts.
• 10:05 AM: The user browses your product and pricing pages.
• 10:10 AM: They remember an email from you with a special offer, find it in their inbox, and click the link. They land on a special offer page (Source/Medium = newsletter / email).

In Universal Analytics, this was recorded as two separate sessions. The first session was attributed to Google Ads and ended at 10:10 AM. The second session started immediately, attributed to your email newsletter. This approach often inflated session counts and made it difficult to see how different marketing channels worked together to influence a single visit.

Understanding Sessions in Google Analytics 4

GA4 completely rethinks this process with its new event-based data model. In GA4, a session is defined simply as a group of events a user triggers within a specific timeframe.

Crucially, GA4 does not start a new session when the traffic source changes mid-visit.

Once a session starts, it only ends after a period of inactivity (the default is 30 minutes, but this is customizable). That's it. No campaign overrides, no midday resets.

Let’s run through that same user journey, but this time see how GA4 records it:

• 10:00 AM: User clicks a Google Ad (Source = google, Medium = cpc) and lands on your homepage. GA4 triggers a session_start event. A single session begins, with its attribution set to 'google / cpc'.
• 10:05 AM: The user browses pages, triggering page_view events.
• 10:10 AM: They click the link from the email newsletter (Source = newsletter, Medium = email). This simply triggers another page_view event within the same session.

In GA4, this entire journey is recorded as one continuous session. The session's primary source and medium remain what brought the user in initially: 'google / cpc'.

AI Agents for Marketing

Crash Course

Graphed

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Download the PDF

What's with the 'session_start' Event?

Since everything in GA4 is an event, the concept of a "session" is handled by a special, automatically collected event called session_start. Whenever a user begins a new visit to your site after a period of inactivity, GA4 logs this event.

This event carries parameters that help Google group all subsequent events together, like:

• ga_session_id: A unique timestamp that identifies the specific session.
• ga_session_number: A number that counts how many sessions this specific user has started.

Think of session_start as the official starting pistol. Once it fires, GA4 considers all subsequent events from that user part of the same "race" - until they stop running for 30 minutes.

What This Change Means for Your Marketing Reports

This fundamental shift from UA's session logic to GA4's method might seem subtle, but it has significant implications for how you analyze your marketing performance.

1. More Accurate User Journey Insights

The biggest benefit is a more realistic view of user behavior. Users rarely interact with a business in a straight line. They get distracted, check other sources, and come back. GA4's model reflects this messy, multi-touch reality.

You’ll likely notice that your GA4 reports have a lower session count than your old UA reports for the same period. This is a good thing! It’s not because you have less traffic, it’s because GA4 is no longer artificially splitting single visits into multiple sessions. You're getting a more accurate, de-duplicated count of user engagement.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

2. Understanding Session-Based Attribution

This is where marketers need to adjust their thinking the most. In GA4, the session source / medium dimension shows you what started the session. It does not change halfway through.

In our example, if the user went on to make a purchase at 10:15 AM, the Acquisition reports in GA4 would credit that session's conversion to 'google / cpc' because that channel initiated the visit. The email click is seen as an influencing touchpoint within that session, not the start of a new one.

This shift emphasizes analyzing the "session starter" and using different attribution models for a more holistic view of which channels influence conversions, rather than just which channels started a particular session.

3. Meet 'Engaged sessions'

With the old session-splitting logic gone, GA4 introduced a more meaningful metric than Bounce Rate: Engaged sessions. An engaged session is a visit that meets any of the following criteria:

• Lasts longer than 10 seconds (this duration is adjustable).
• Includes a conversion event.
• Has at least two pageviews or screenviews.

This metric helps you quickly separate visitors who showed some intent from those who left immediately, providing a much clearer signal of content and channel quality.

But What If I Need to Know About That Mid-Session Source Change?

Reading this, you might be thinking, "That's great, but the newsletter click in that session is important! How can I see its impact?"

You absolutely can. While GA4 doesn't create a new session, it still tracks the source information for every event. You just need to look for it in the right place, typically moving from standard reports to Explorations.

AI Agents for Marketing

Crash Course

Graphed

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Download the PDF

Look at Event-Level Source Data

In generating reports, you're looking at data that is aggregated up to the session or user level. But GA4 gives you the power to go deeper.

By creating a custom Exploration report, you can analyze events directly. For example, you could create a table that shows page_view events and pair it with an event-scoped traffic source dimension. This would allow you to see that while the 'google / cpc' session was in progress, a specific page view event came from the 'newsletter / email' source.

Use GA4's Attribution Modeling

The standard Acquisition reports primarily focus on what drove the session. To understand the influence of touchpoints like the mid-session email click, you should look at the attribution reports under 'Advertising'.

GA4's default attribution model is data-driven, which assigns credit to various touchpoints throughout the entire customer journey based on their contribution to a conversion. In this model, both the initial Google Ad and the later email newsletter click would likely receive partial credit for an eventual sale, giving you a much more nuanced view of what works.

Final Thoughts

Switching from Universal Analytics to GA4 requires rethinking some core concepts, and the definition of a "session" is one of the biggest changes. GA4’s decision to stop creating new sessions when a campaign source changes provides a more accurate, user-centric view of an entire visit, preventing inflated metrics and clarifying the customer journey.

Analyzing the complete journey across GA4, Google Ads, Shopify, and your CRM can feel like a full-time job of manual exporting and data wrangling. At Graphed, we created our AI analyst to fix this. Just connect your platforms and use plain English to build real-time marketing dashboards instantly - no more wrestling with CSV files or trying to remember which GA4 report shows you what you need. We help you get straight to the insights so you can get back to growing your business.