Google Analytics 4: Tips and Tricks to Get the Most Out of Your Data

April 2021

Welcome to the 2nd update in the series. For those who missed out on the first update or need a recap, have a look at Google Analytics 4: The Business and Future Focused Analytics Tool and then come back.

This time, we would like to focus on the following abilities gained through GA4:

  • Trusting your data and avoiding incorrect traffic attribution
  • Retaining data: Do you need access to historical data?

These abilities can be gained through two of the settings which require special attention during configuration:

  • Exclusion of Unwanted Referrals
  • Data Retention settings

We would also like to introduce Data Deletion requests, which – in the privacy-sensitive age that we live in – could provide an escape route from an unintentional privacy breach in Google Analytics.

Not sure how to get started with reporting on your GA4 data? Sign up for our newsletter to stay up to date, or

GA4 Newsletter Series #2: Trust Your Data and Improve Data Quality

(Self-)Referral exclusion in GA4 – Significantly improve your data quality

Data quality has always been one of the hot topics of the industry, with one major question often posed by digital marketers: “How can we be sure we can trust this data?” The building of that trust requires extra care at the time of data collection. And one of the most common pitfalls is incorrect attribution of traffic (and conversions).

Scenario 1

Imagine a visitor arriving at your website from a newsletter campaign. They browse the website and add some items to the cart, but before making that purchase they log into your website using Facebook. And so, the last traffic source recorded in Google Analytics is… yes, that’s correct: Facebook. As a result, Google Analytics attributes the purchase to Facebook, not the newsletter. Would you trust that data?

Scenario 2

As happens often, the visitor chooses to make a payment using a debit card, or another 3rd party payment platform. Once the payment has processed, they are redirected to the thank-you page on your website. Again, Google Analytics will attribute that data to the most recent 3rd party site: the payment platform. Isn’t that just wrong?

Luckily, Google Analytics 4 already offers a solution to incorrect traffic attribution: Exclusion of Unwanted Referrals.

In order to make sure that Google Analytics does not attribute your traffic to 3rd party websites such as login or payment platforms, go to Settings > Data Streams. Next, select your data stream and under More Tagging Settings > Additional Settings click List Unwanted Referrals. You will then be able to specify which domains you do not want to include as traffic sources in your Google Analytics reports.

With that setting in place, Google Analytics will “skip” the specified websites in its calculation of sessions. As a result, your reports will correctly display your campaign – such as newsletter – as the data source responsible for the conversion:

And last but not least…

Scenario 3

Perhaps less common but similarly misleading are self-referrals in Google Analytics. In this scenario, the user could arrive at an incorrectly tagged landing page, where the Google Analytics tracker does not fire. We have come across this behaviour on websites with incorrectly implemented cookie consent platforms. As a result, your session would be attributed to your own landing page:

Luckily, in the case of self-referrals, Google Analytics 4 will by default attribute your traffic correctly. We by all means discourage erroneous implementations that could cause such behaviour, but – while your development teams are working to identify the issue – at least Google Analytics will be recording the correct data sources:

Can we trust this data? Much more so. With greater control over traffic attribution, Google Analytics 4 will report correct campaigns as sources of your conversions.

Data deletion requests in GA4 – Delete PII data in your property

Data deletion is probably one of the lesser known Google Analytics features. It comes in handy though, when you need to delete data from the GA4 servers for any reason.

Currently, we see 2 major use cases in your day-to-day work:

  • Deletion of accidentally tracked personal identifiable information (PII)
  • Deletion of unwanted campaign data

Assume you have accidentally been collecting credit card data or email addresses of your users in the page path. Now, you must get rid of these data points to stay compliant with GA’s Terms of Service and protect your users’ privacy. First, you have ensured that this data is no longer collected by updating the Google Tag Manager tags, but you are left with the historical data in your GA properties.

Data deletion requests allow you to precisely delete these unwanted data points for any given period in GA4.

The feature is much more flexible in GA4 than it used to be in Universal Analytics. Instead of Universal Analytics’ “all-or-nothing” approach, you can be incredible flexible and limit the deletion request to specific:

  • Periods of time
  • Events
  • Event parameters (that contain certain values)
  • User properties (that contain certain values)

We are happy with the updated data deletion feature as you truly get to choose what data points to keep and what to delete depending on your organization’s requirements. It is a significant improvement compared to Universal Analytics and yet another example of the flexibility that we gain by adopting GA4’s event-based data model.

Data retention in GA4 – Ensure to keep your data

Data retention is a GA4 setting that strongly determines your data availability. With the data retention control, you can specify the amount of time before Google will delete all user- and event-level data from the GA4 servers. Think of event-level data as your website traffic at the most granular level – a single hit with all its characteristics.

The default data retention is two months, but you can set it to 14 months (maximal value).

Most of GA4’s pre-built standard reports, however, are based on aggregated data. Aggregated data is the result of condensing your hit-level website traffic along certain dimensions. An example would be the sum of all sessions (aggregated metric) per channel (dimension). 

These kinds of reports are not affected by the data retention settings and, therefore, will always be available for analysis. Even long date comparisons will work just fine in GA4’s standard reports.

So, why care about the data retention settings then? If you have been working with GA4 already, you might have noticed that GA4’s standard reporting is minimal compared to good old Universal Analytics. To overcome these limitations, GA4 makes wide use of the Analysis hub.

For the Analysis hub, GA4 uses fine-grained, event-level data to give you all the flexibility you need to answer specific questions. However if your data retention settings are currently two months and you want to analyze data for the last six months, you will run into a similar error like this:

If you were used to the 26-month limit on UA and the new 14-month limit is too restrictive, you still have the possibility to activate the BigQuery export, which will allow you to save a full history of your data at user and event level. 

To summarize, if you do not plan to use anything beyond standard reports, there are no changes you have to make. However, if you are interested in reporting user and event level data over long periods, you should extend the default retention to 14 months. Furthermore, if the 14 months is too short or you are interested in segmenting users or predicting their behaviour, you should activate the BigQuery data export right away to guarantee the availability of historical data!  

Closing thoughts

Within this edition of the GA4 newsletter series we shed light on a few features that are critical when you configure your GA4 properties for the first time. Checking your domain configuration and data retention settings are definitely worth your time and your colleagues will thank you for it later. The referral exclusion as well as the data deletion features of GA4 are another good example of how much more flexible GA4 is compared to Universal Analytics.

Stay tuned as we have more insights and tips and tricks in the making!

As always – let us know if you have any feedback or want to know more about a specific GA4 feature. Simply drop us a line at or if we can help you further.   |   |   +45 70 20 29 19

Florian Pertynski

Florian Pertynski

Senior Analytics Specialist, IIH Nordic
Gunnar Griese

Gunnar Griese

Analytics Specialist, IIH Nordic