Google Analytics 4
Google Analytics 4 (GA4) is Google current web and app analytics platform. It replaced Universal Analytics in 2023 with a fundamentally different data model based on events rather than sessions. For marketing technology professionals, GA4 is the foundation of digital measurement: it tracks user behavior across websites and apps, provides audience insights, and feeds data to advertising platforms for optimization.
This page covers what GA4 is, how I implement and use it, common mistakes I see, and how it fits into the broader marketing measurement stack.
What It Is and Why It Matters
GA4 is built on an event based data model. Everything is an event: page views, clicks, form submissions, purchases, scroll depth. This is different from the session based model of Universal Analytics. The event model is more flexible and better suited to tracking user journeys across devices and platforms.
Key capabilities include: cross device and cross platform tracking (web and app in one property), machine learning powered insights and predictions, BigQuery integration for raw data access, enhanced measurement for common events without custom code, and privacy focused design with consent mode and data deletion controls.
For marketing technology teams, GA4 serves as the central data source for understanding digital behavior. It integrates with Google Ads for conversion tracking, with Google Tag Manager for event implementation, and with BigQuery for advanced analysis. Getting GA4 right is critical because nearly every marketing decision depends on the data it provides.
Common Use Cases
What GA4 is used for in practice.
Marketing Performance Measurement
Tracking the effectiveness of marketing campaigns across channels. Understanding which sources drive the most valuable traffic and conversions. Providing the data foundation for [attribution modeling](/attribution-modeling-explained).
Audience Analysis
Understanding who visits your site and app, what they do, and where they drop off. Building audience segments based on behavior for remarketing and personalization.
Conversion Tracking
Defining and tracking key business events like purchases, sign ups, and lead form submissions. Sending this conversion data back to advertising platforms for campaign optimization.
User Journey Analysis
Mapping how users navigate through your site or app across multiple sessions and devices. Understanding the paths that lead to conversion and where users abandon.
Raw Data Export to BigQuery
Exporting event level data to BigQuery for custom analysis, machine learning, and integration with other data sources. This is where GA4 becomes truly powerful for data driven marketing.
Privacy Compliant Analytics
Implementing analytics that respect user consent preferences using Consent Mode. Collecting data in a way that complies with GDPR and other privacy regulations while still providing useful insights.
Practical Experience
I have implemented GA4 for companies ranging from small businesses to enterprise organizations. My experience covers the full implementation lifecycle: planning the measurement strategy, configuring data streams, implementing custom events through Google Tag Manager, setting up conversions, and building reporting in Looker Studio.
The most important lesson is that GA4 implementation quality directly determines the quality of every marketing decision. If events are tracked inconsistently, if conversion definitions do not match actual business outcomes, or if consent mode is not implemented correctly, then reports are misleading and advertising platforms optimize for the wrong things.
I work extensively with the GA4 to BigQuery export, which provides access to raw event level data. This unlocks analysis that is impossible in the GA4 interface: custom attribution models, cohort analysis with SQL, customer lifetime value calculations, and integration with CRM data. For companies serious about data driven marketing, the BigQuery integration is where GA4 delivers its greatest value.
In the French market, GA4 implementation requires particular attention to consent mode. CNIL requires explicit consent before tracking cookies are set, which means GA4 must be configured to respect consent signals. I implement consent mode with modeling, which allows GA4 to estimate overall behavior even when some users do not consent, maintaining data quality within regulatory requirements.
Common Mistakes to Avoid
Common GA4 implementation mistakes that undermine marketing measurement.
No Measurement Plan
Implementing GA4 without first defining what to measure and why. Start with business questions, define the events and conversions needed to answer them, then implement. Without a plan, you collect data nobody uses.
Incorrect Event Configuration
Setting up custom events with inconsistent naming, missing parameters, or wrong scopes. This makes data unreliable and reporting impossible. Follow a clear naming convention and test every event before going live.
Ignoring Consent Mode
Not implementing consent mode in regions that require it. This creates both regulatory risk and data quality issues. Consent mode with modeling is the correct approach for GDPR markets.
Wrong Conversion Definitions
Marking the wrong events as conversions. If a page view or button click is marked as a conversion instead of an actual purchase or lead, all advertising optimization based on GA4 data will be wrong.
Not Enabling BigQuery Export
Missing the opportunity to export raw data to BigQuery. This is free for GA4 properties and provides capabilities that the GA4 interface cannot match. Enable it early to start building historical data.
Frequently Asked Questions
The standard version of GA4 is free. GA4 360 (the enterprise version) is paid and offers higher data limits, more custom dimensions, and SLA backed support. Most companies can operate effectively on the free version. The BigQuery export, which is one of GA4 most powerful features, is available on the free version (you pay only for BigQuery storage and queries, which is typically very low cost).
A basic implementation with standard events can be done in one to two weeks. A comprehensive implementation with custom events, enhanced e-commerce tracking, consent mode, BigQuery export, and Looker Studio reporting typically takes four to eight weeks. The planning phase (defining what to measure) is as important as the technical implementation.
GA4 is the default choice for most companies because of its integration with the Google ecosystem (Google Ads, BigQuery, Looker Studio). Alternatives like Adobe Analytics, Amplitude, or Mixpanel may be better for specific use cases (app analytics, product analytics). For marketing measurement specifically, GA4 combined with BigQuery is very hard to beat on value for money.
GA4 includes consent mode (adjusting data collection based on user consent), data deletion controls, IP anonymization by default, and configurable data retention periods. It can be implemented in a privacy compliant way, but it requires proper configuration. Simply adding the GA4 tag without consent management is not compliant in GDPR markets.
Events track user actions (page views, clicks, form submissions). Conversions are specific events that you mark as important business outcomes (purchases, lead form completions, sign ups). You can mark any event as a conversion in GA4. This distinction matters because advertising platforms optimize toward conversions, so choosing the right events to mark as conversions directly affects campaign performance.
Related Topics
Marketing Technology
MarTech strategy.
Google Tag Manager
Tag management.
BigQuery
Data warehouse.
SQL for Marketing
Data analysis.
Marketing Technology Manager
MarTech leadership.
Marketing Analytics Manager
Analytics leadership.
Server Side Tracking
Server side data collection.
What Is Marketing Technology
MarTech explained.
Need Help with Google Analytics 4?
I implement GA4 properly: measurement planning, event configuration, consent mode, BigQuery export, and actionable reporting.