How To Master GA4: Benefits, Migration and Alternatives

• Pros & Cons of GA4
• Differences Between UA and GA4
• Data Challenges
• BigQuery, The Solution To Data Challenges
• GA4 Account Setup Framework
• GA4 Alternatives

Thresholding places limits on the number of rows and dimensions in reports, potentially limiting the granularity and flexibility of reporting for some businesses. There are some mitigations for this, but you will lose the ability to use advanced audience features in reporting and Google Ads.

Cardinality causes similar issues as thresholding and is where Google groups data with a high number of dimensions together in order to reduce the processing cost. This can result in up to 70 percent of rows being reported as (other)

Integrating GA4 data with other platforms or tools, such as Looker Studio, may present some API limitations that limit data included in reporting, making it impossible for businesses that rely on these integrations for their analytics needs.

GA4 has a more complex interface and data model, which may make it harder for businesses with limited technical expertise to navigate and understand the platform compared to Universal Analytics.

For example, a source / medium report doesn’t come out of the box with a drill down. The customizable reporting interface is lacking many features that would allow you to drill down into drivers of trends. Reports can be shared with others, but only the original report creator can change settings, or even the time period for the report, rendering this feature useless.

There's no ability to import your UA data into GA4.

Data Model Changes

• One of the primary differences between Universal Analytics (UA) and GA4 is the data model. GA4 focuses on an event-driven data collection approach, which “enables businesses to track user behavior more accurately and comprehensively”.
• This new data model allows for enhanced measurement capabilities for user engagement, largely by making the computing time of machine learning features faster, cheaper, and more integrated with other Google Marketing Platform products (like Google Ads).
• GA4 uses an event-based model to track user interactions, which is a departure from the session and pageview-centric approach of Universal Analytics.
• This event-based tracking allows for more granular and customizable tracking of user actions, providing businesses with a wealth of information on user behavior that can be used to optimize their digital presence.
• With GA4, companies can define and track custom events, enabling them to gain a deeper understanding of their users' activities and preferences. The only problem is, the more customizations you add, the faster your data will reach thresholding limits. dohh!

Data Challenges

There are four primary concerns when it comes to data challenges:

• Thresholding
• Cardinality
• Having to build your reporting interface
• API Limitations (with Looker Studio especially)

Thresholding

GA4 introduces two HUGE data challenges that businesses should be aware of before migrating. Thresholding and Cardinality.

Thresholding is applied to your GA4 account whenever individual users might be identified by the data shown in a report. Thresholding generally happens when reports segment data with dimensions that include demographic data or other affinity attributes.

Data thresholding in Google Analytics 4 should not be confused with data sampling, as they are distinct concepts. Thresholding is applied to your GA4 property when all the following conditions are met:

• You have collected data through Google Signals by enabling it at some point.
• Your reporting identity is either Blended or Observed.
• A report you're viewing includes rows with low user, event, or session numbers (around 50 or so)

When these conditions are met, rows with low numbers will be hidden and not displayed in the report, although the data remains available in the background. To prevent data thresholding in the future, you can avoid enabling Google Signals if you don't intend to use remarketing features or demographic reports in GA.

Cardinality

Cardinality causes similar issues as thresholding and is where Google groups data with a high number of dimensions together in order to reduce the processing cost. This can result in up to 70 percent of rows being reported as (other). Imagine looking at your traffic channels report and seeing that 70% of your data is grouped in “Other” even though your traffic channels are setup correctly! Have lots of custom events or parameters? This will quickly limit the granularity and flexibility of reporting, making it difficult for businesses to access and analyze all of the data they need.

Reporting Interface

With GA4, businesses may need to build their reporting interface to accommodate custom reports and advanced analysis requirements. This can potentially increase the complexity and resource requirements of the migration process, as businesses may need to invest in additional development resources to create and maintain these custom reporting interfaces. Here’s some examples of GA4 reports that don’t exist or function the same way as UA:

1. Channels report (no drill down)
2. Source / medium
3. Landing Pages

Yes, we were surprised by these core missing reports too!

API Limitations

Finally, integrating GA4 data with Looker Studio may present some API limitations that could impact reporting and analysis capabilities. To be specific, with GA4, Google started actually enforcing quotas between Looker Studio and Google Analytics. These quotas always existed, they just weren't enforced for UA data to Looker Studio. Businesses relying on Looker Studio for their analytics needs should carefully consider these limitations and weigh the benefits of GA4 against the potential challenges they may face in integrating the two platforms.

Read about the actual quota limits here.

BigQuery Is The Solution To Data Challenges

BigQuery can help address some of the data challenges that you may face when migrating to GA4. By integrating GA4 with BigQuery, you can access raw data for advanced analysis and reporting. Just get a nice reporting tool to do it: Tableau, Looker, PowerBI, Domo, Etc.

This can help bypass the thresholding and privacy limitations that may be imposed by GA4's built-in reporting features. Not to mention you have complete data ownership and can integrate it with any of your other business applications.

With BigQuery, businesses can perform in-depth analysis and create custom reports that cater to their specific needs, without being constrained by GA4's default reporting capabilities. Exporting your data to BigQuery is also the first step to linking analytics data to other data, running forecasts with web data, using web data to feed back into CRM systems, and finally, data science. The question to ask is, do you have the skills to do this internally? Or do you need to find a partner who can help do this for you?

Set Up The BigQuery Integration In GA4

To set up the BigQuery integration with GA4, connect your GA4 property to a BigQuery project. Once connected, GA4 will automatically export raw data to BigQuery, which can then be processed and analyzed.

To use this raw data in reporting tools like Tableau, you will need to process the data, transforming it into a “model” for visualization and analysis. The raw data GA4 puts into BigQuery isn’t transformed to be used in reporting, and requires a fair amount of work to aggregate and calculate metrics, plus build reporting tables.

This may involve writing custom queries and scripts to extract, transform, and load the data into Tableau or other reporting tools. It is a bit of a pain. The good news is, our team can make it stress-free and do it for you by deploying custom reporting tables we’ve created based on the Google Analytics 4 data model.

Remember leveraging BigQuery's powerful data processing capabilities is what allows you to get past some of the limitations of GA4.

Configurations:

1. Configure data streams for websites, apps, or both.
2. Set up user properties, such as user demographic data, to store custom user-level data.
3. Configure event parameters for custom events, such as button clicks or form submissions.
4. Set your most important events as goals.
5. Set up custom dimensions and metrics to track additional data points specific to your business.
6. Configure data retention settings to manage how long user data is stored in GA4.
7. Implement consent settings to ensure GDPR, CCPA, and other data protection regulation compliance.

Implementations:

1. Implement GA4 tracking code (gtag.js or Google Tag Manager) on your website or mobile app to start collecting data.
2. Set up event tracking for specific user interactions, such as clicks, page views, or form submissions, by modifying the tracking code or using Google Tag Manager.
3. Implement Enhanced Measurement to automatically track key user interactions without additional code.

Integrations:

1. Integrate GA4 with Google Ads to track conversions and optimize ad campaigns.
2. Set up Google Tag Manager to manage tracking codes and events more easily.
3. Server side tracking recommended.
4. Integrate GA4 with BigQuery to store and analyze raw data.
5. Connect GA4 with third-party tools, such as CRM systems or marketing automation platforms, to enable data sharing and improve overall business processes.
6. Consider any migrations for reporting you may have to make for existing reporting or tools hooked up with UA.

Executive

• Looking for how most important KPIs are aligning to business goals such as: revenue, conversion rates, and customer acquisition costs.

Marketing

• Wants to have all the data available at all times and then never use it (#satire)
• Plan, execute, and optimize marketing campaigns across various channels.

IT

• Site speed
• Tagging & reliability
• Audits of integrations with tracking code

Sales

• Understanding of what moves customers through each step of the sales funnel.

Product

• Feature engagement

RFM analysis

• Use this to devise a list and requirements for what reporting you’ll need to develop.

Alternative Platforms

The following are the list of alternative platforms we would recommend to clients:

1. Adobe Analytics
2. SnowPlow
3. Matomo (formerly Piwik)

Let’s quickly discuss each one below.