GA4

GA4 is the default analytics layer for almost every marketing team on the web. It tracks how people find the site, what they do once they arrive, and which actions ladder up to revenue. The platform is free for most companies, deeply tied to Google Ads, and exports raw event data to BigQuery out of the box. The problem is that most installs are surface-level. Teams paste the tag, accept the defaults, and miss almost everything that makes GA4 interesting.

What GA4 Does

GA4 replaced Universal Analytics in July 2023 and rebuilt the data model from the ground up. Every interaction is an event, every event can carry parameters, and every user is stitched across devices using a combination of Google Signals, user ID, and modeled identity. That shift is what unlocks the modern reporting and automation playbook.

• Event-based tracking with up to 25 custom parameters per event, replacing the rigid pageview and session model from Universal Analytics.
• Cross-platform measurement that unifies web and mobile app behaviour into a single property, with Firebase as the native mobile SDK.
• Native BigQuery export on the free tier (1M events per day standard, billions for GA4 360), giving every team access to raw event data.
• Data-driven attribution as the default model, using machine learning to credit conversions across the full path rather than last click.
• Audience builder that creates behavioural segments and syncs them to Google Ads, Display & Video 360, and Search Ads 360 in near real time.
• Consent Mode v2 and Google Signals support, which model conversions for users who decline tracking while keeping the install compliant.
• Server-side Google Tag Manager integration for first-party data collection that survives browser restrictions and ad blockers.

GA4's AI

GA4 ships with a layer of machine learning under the hood. Predictive metrics like purchase probability, churn probability, and predicted revenue are calculated for every user with enough history, and they can be used directly in audience definitions. Automated Insights surface anomalies and trends in the Reports view, and the natural-language search bar in the UI lets non-technical operators ask questions like "top campaigns by conversion last month" without building a custom report. When third-party cookies disappear or consent is denied, behavioural and conversion modeling fills the gaps so the numbers stay usable.

Automations We Build with GA4

GA4 is most valuable when it stops being a reporting tool and starts being the conversion data backbone for the rest of the stack. These are the builds we run for mid-market teams that want GA4 to do real work, not just show last week's traffic chart.

• Server-side GTM rollouts that route every event through a first-party endpoint, recover 10 to 30 percent of conversion data lost to ad blockers, and feed Enhanced Conversions into Google Ads.
• BigQuery pipelines that merge GA4 raw events with Stripe, HubSpot, Salesforce, and ad spend so finance can pull a true cost per acquisition by source on demand.
• Audience syncs into Google Ads, Meta, and LinkedIn that target high-intent behavioural segments (cart abandoners, pricing page visitors, demo no-shows) for retargeting and exclusion.
• Slack alerts triggered by GA4 anomalies or BigQuery scheduled queries when conversion rate, funnel completion, or core event volume drops outside a defined range.
• Custom Looker Studio or Power BI dashboards on top of the BigQuery export, replacing the slow default GA4 explorations with reports that load in seconds.
• Lead scoring models that combine GA4 behavioural signals (sessions, content depth, pricing visits) with CRM data to push hot accounts into the SDR queue.
• Consent Mode v2 deployments with proper basic and advanced signals so EU traffic still produces modeled conversions instead of going dark.

Why Teams Choose GA4

• Free at the standard tier with limits most mid-market sites never hit, including 10 million events per month on standard properties and unlimited streaming export to BigQuery.
• Native integration with Google Ads, Search Ads 360, and Display & Video 360, which is the shortest path to conversion-aware bidding and audience targeting.
• Raw event export to BigQuery means the data is yours. Once it lands, you can join, query, and model it without waiting for the GA4 UI to render.
• Consent and privacy primitives are built in, including Consent Mode v2, IP anonymisation, and modeled conversions for users who opt out.
• Predictive audiences and data-driven attribution come standard, no separate licence required, which used to be a six-figure Adobe Analytics ask.

GA4 integrates with Google Tag Manager, BigQuery, Looker Studio, Google Ads, Search Console, Firebase, Salesforce Marketing Cloud, HubSpot, Segment, and almost every server-side data platform on the market. The standard tier is free; GA4 360 (the enterprise SKU) starts around $50,000 per year and unlocks higher limits, sub-property and roll-up support, longer retention, and an SLA. For most mid-market companies, free GA4 plus a proper server-side install and BigQuery pipeline beats GA4 360 on cost and on what actually matters. That is the build we do.