Google Analytics 4 (GA4) officially replaced Universal Analytics (UA) last year, but there’s still a lot to get used to. Even if you’ve been collecting data in GA4, the way that data is collected and displayed has changed in ways that can still surprise users. Here are five tips to make sure that you get the data you need and can understand it correctly.
1. Change the data retention time
Data retention is how long Google Analytics will hold onto information. With an increased focus on web privacy across the globe, Google has implemented two settings for data retention. Your GA4 will either retain data for two months or 14 months.
We recommend changing the data retention setting to 14 months, especially for our higher ed clients. This setting will only affect explorations (a new form of reports in GA4), but you want the ability to compare data year-over-year. Two months of data just isn’t enough to make robust explorations.
You’ll want to change this setting as soon as possible. You won’t be able to see any data older than two months, but you will stop deleting data so you will have access to all of the information in the future.
2. Review session timeout settings
The session timeout determines when to start a new session. When a browser tab or window sits idle for too long, Google Analytics considers the current session over. When you click back to the tab or window, the timer starts again and triggers a new session instead of continuing the previous one.
Depending on your business, you may want to adjust the length of a session. GA4 defaults to 30 minutes. If you send users to another site for applications or other resources, you may want to extend the timer to an hour.
3. Understand engagement in your GA4 implementation
GA4 measures engagement differently, but it’s not obvious how unless you dig into the topic. In UA, engagement was based on how many pages a user saw. Analytics used to display Bounce Rate to show how many users weren’t visiting more than one page.
In GA4, Google implemented a more nuanced approach to engagement. Rather than number of pages, a session is now labeled engaged based on either:
- If the user stays on the page for a set number of time
- If the user interacts with an element on the page
There’s a new timer that starts when a user visits a page to determine whether a session qualifies as engaged. GA4 defaults to a 10-second timer, which doesn’t provide much information in practice. It can take longer than 10 seconds for a user to assess if the page has the info they’re looking for or not. Consider adjusting the timer to somewhere between 30 to 60 seconds for engaged sessions to more accurately determine a user’s interest in both the webpage and the site.
4. Don’t look at data in the past 24 hours
Sometimes you just want to know what’s going on with a new page. Maybe you just reworked a section of your site or launched a marketing landing page. GA4 takes time to process data, and if you try to look at that data too quickly, it won’t be very accurate. Specifically, you may notice a spike in “unassigned” traffic. GA4 labels traffic as unassigned first, but will reclassify it as it processes the data. If you want to see the most accurate information, you’ll have to wait the full 24-48 hours Google claims.
5. Know that GA4 conversions are now called key events
In March 2024, Google changed the name of GA4 “conversions” to “key events.” Functionally this doesn’t change anything, but it does mean that a lot of different reports have new names or metrics have been changed. When you see the term “conversion” now, it’s exclusively talking about Google Ads conversions. It should help clarify discrepancies between Ads conversions and GA4 events. Take time to get used to the new terminology, and know that it could be a while until that language is available universally.
Want to know more about making the most of your GA4 data? Contact us to see how we can help you get started.
Posted on May 28, 2024 by TJ Gunther in Commentary
Google Analytics 4 (GA4) officially replaced Universal Analytics (UA) last year, but there’s still a lot to get used to. Even if you’ve been collecting data in GA4, the way that data is collected and displayed has changed in ways that can still surprise users. Here are five tips to make sure that you get the data you need and can understand it correctly.
1. Change the data retention time
Data retention is how long Google Analytics will hold onto information. With an increased focus on web privacy across the globe, Google has implemented two settings for data retention. Your GA4 will either retain data for two months or 14 months.
We recommend changing the data retention setting to 14 months, especially for our higher ed clients. This setting will only affect explorations (a new form of reports in GA4), but you want the ability to compare data year-over-year. Two months of data just isn’t enough to make robust explorations.
You’ll want to change this setting as soon as possible. You won’t be able to see any data older than two months, but you will stop deleting data so you will have access to all of the information in the future.
2. Review session timeout settings
The session timeout determines when to start a new session. When a browser tab or window sits idle for too long, Google Analytics considers the current session over. When you click back to the tab or window, the timer starts again and triggers a new session instead of continuing the previous one.
Depending on your business, you may want to adjust the length of a session. GA4 defaults to 30 minutes. If you send users to another site for applications or other resources, you may want to extend the timer to an hour.
3. Understand engagement in your GA4 implementation
GA4 measures engagement differently, but it’s not obvious how unless you dig into the topic. In UA, engagement was based on how many pages a user saw. Analytics used to display Bounce Rate to show how many users weren’t visiting more than one page.
In GA4, Google implemented a more nuanced approach to engagement. Rather than number of pages, a session is now labeled engaged based on either:
There’s a new timer that starts when a user visits a page to determine whether a session qualifies as engaged. GA4 defaults to a 10-second timer, which doesn’t provide much information in practice. It can take longer than 10 seconds for a user to assess if the page has the info they’re looking for or not. Consider adjusting the timer to somewhere between 30 to 60 seconds for engaged sessions to more accurately determine a user’s interest in both the webpage and the site.
4. Don’t look at data in the past 24 hours
Sometimes you just want to know what’s going on with a new page. Maybe you just reworked a section of your site or launched a marketing landing page. GA4 takes time to process data, and if you try to look at that data too quickly, it won’t be very accurate. Specifically, you may notice a spike in “unassigned” traffic. GA4 labels traffic as unassigned first, but will reclassify it as it processes the data. If you want to see the most accurate information, you’ll have to wait the full 24-48 hours Google claims.
5. Know that GA4 conversions are now called key events
In March 2024, Google changed the name of GA4 “conversions” to “key events.” Functionally this doesn’t change anything, but it does mean that a lot of different reports have new names or metrics have been changed. When you see the term “conversion” now, it’s exclusively talking about Google Ads conversions. It should help clarify discrepancies between Ads conversions and GA4 events. Take time to get used to the new terminology, and know that it could be a while until that language is available universally.
Want to know more about making the most of your GA4 data? Contact us to see how we can help you get started.
Tagged With These Topics
About TJ Gunther
Related Posts
Any Way You Want It
Post by Bruce Cronlund
The success or failure of your CMS is going to depend much more on how that CMS is implemented than in the capabilities of the CMS itself.
The ROI of Time: Whose Is More Valuable, Yours or Your User’s?
Post by Lamar Goodenough
How do we transform reams of complicated requirements from legacy print formats into strong and nimble web content?