Analytics & Tracking

GA4 Configuration & Events
Make GA4 Actually Useful for Revenue.

GA4 property setup, data streams, custom events, key events, explorations, audiences and BigQuery export. We configure GA4 so leadership and marketing get clear answers instead of confusing reports that nobody trusts.

Get a Free GA4 Audit arrow_forward
↑ 4.2×
Actionable Insights
96%+
Event Coverage
↓ 60%
Time to Answer
85+
GA4 Properties Configured

Why Default GA4 Is Rarely Enough

The Default Problem

  • Only enhanced measurement events with almost no business context
  • Key events that do not match actual revenue or pipeline stages
  • No custom dimensions for the things that actually segment your customers
  • Attribution models left on default last-click
  • Explorations that nobody on the team knows how to use

Our Philosophy

GA4 is a raw event stream and a powerful analysis engine. We treat it as both. We define the events and parameters that map to your funnel, wire the right custom dimensions and metrics, configure meaningful explorations and audiences, and connect BigQuery when you need to go beyond the UI.

Deep GA4 Configuration Details

Property & Stream Architecture

  • Data Streams: Web + iOS + Android when relevant, with correct measurement IDs and Firebase linkage for apps.
  • Subproperties & Rollups: When you have multiple brands or regions that need both consolidated and separate views.
  • Data Filters & Settings: Internal traffic exclusion, developer traffic, data retention, and session timeout tuned to your actual buyer behavior.

Events, Parameters & Dimensions

  • Key Events: Properly marked revenue events (purchase, subscribe, qualified lead) with value, currency, and transaction id.
  • Custom Dimensions: User-scoped (customer type, account tier), Event-scoped (lead source detail, campaign variant), and Item-scoped for ecommerce.
  • Enhanced Measurement + Custom: We keep what is reliable and add the custom events that actually drive decisions (demo requested, pricing viewed, account created, etc.).
  • BigQuery Export: Enabled with partitioned tables, scheduled queries, and views that marketing and leadership can actually use.

Our GA4 Configuration Process

01

Current Property & Data Audit

Review streams, events, parameters, audiences, explorations, attribution settings, and data quality issues in the actual reports.

02

Measurement Strategy Workshop

Map your funnel stages, micro-conversions, and the dimensions that actually segment performance (channel, campaign, persona, product, etc.).

03

Property & Stream Hardening

Correct data stream config, internal traffic rules, session settings, data retention, and consent-aware collection.

04

Event & Parameter Implementation

Define the full event catalog in GTM or gtag. Add custom dimensions/metrics. Wire value, currency, and item data where revenue matters.

05

Explorations, Audiences & Attribution

Build the funnel, path, and segment explorations you will actually use. Configure data-driven attribution and predictive audiences when data volume supports it.

06

BigQuery + Governance

Enable and structure export. Create useful views. Deliver documentation so the team can maintain and extend the setup.

Frequently Asked Questions About GA4 Configuration & Events

Technical answers about making GA4 produce numbers you can act on.

What is the difference between events and key events in GA4? expand_more
All events are collected. Key events are the ones you explicitly mark as important for reporting and optimization (formerly conversions). We mark purchase, qualified lead, subscription, and other revenue or pipeline events as key so they appear in key event reports and can be used for bidding.
How do you decide which custom dimensions to create? expand_more
We only create dimensions that will be used for segmentation or filtering in explorations or audiences. We ruthlessly prioritize the 8-12 that actually move decisions instead of creating 60 dimensions nobody uses.
Should we use data-driven attribution or stick with last non-direct click? expand_more
Data-driven when you have sufficient conversion volume (Google recommends ~3000 conversions in 30 days for the model to be reliable). For lower volume we often use position-based or time-decay and document why.
How do you handle purchase events with variable order values and multiple items? expand_more
We implement the full Google ecommerce spec: value, currency, transaction_id, items array with item_id, item_name, item_category, price, quantity. We also add custom parameters like coupon, shipping_tier, and payment_type when relevant.
Can you export everything to BigQuery and build our own dashboards? expand_more
Yes. We enable the export, set up partitioned tables, and create dbt or scheduled query views that marketing and finance can actually use. We also document the schema and common query patterns.
What about user properties vs event parameters? expand_more
User properties for relatively stable traits (customer_type, account_tier, signup_cohort). Event parameters for things that change per interaction (page_template, experiment_variant, lead_score). We keep the distinction clean.
How do you set up cross-device and cross-platform user identification? expand_more
User-ID implementation where you have authenticated users, plus Google Signals where consented. We are careful with consent and regional rules. We also document the resulting identity space limitations.
Do you turn on predictive metrics and audiences? expand_more
Only when the data volume supports stable models. We document the requirements and enable purchase probability, churn probability, and revenue prediction audiences when they are reliable enough to act on.
How do you handle multiple domains or subdomains in one property? expand_more
We use a single property with proper cross-domain configuration (via GTM linker or gtag config) and consistent measurement IDs. We create subproperties or data filters when teams need completely separate reporting.
What explorations do you usually build first? expand_more
Funnel exploration for the main conversion path, path exploration for discovery, segment overlap for audience definition, and free form for the specific questions leadership asks every month.
How do you deal with sampling in GA4 reports? expand_more
We use unsampled explorations when available, BigQuery for anything that needs precision at scale, and we are honest about cardinality and sampling limits in the documentation we deliver.
Can you fix a GA4 property that has been collecting bad data for months? expand_more
We can clean the future and create useful historical views in BigQuery, but past bad data is usually unfixable in the GA4 UI. We document what can and cannot be trusted from the old period.
How do you make sure marketing teams actually use the explorations? expand_more
We build the 5-7 explorations that answer the questions they already ask, record short Loom videos showing how to use each one, and include them in the training session. We do not hand over 40 unused explorations.
What about consent and data retention with GA4? expand_more
We set data retention to the maximum allowed by your consent framework (usually 14 months or 26 months), implement Consent Mode, and document the impact on modeling and reporting for different regions.
How long does a thorough GA4 configuration project take? expand_more
Basic hardening + key events: 2-3 weeks. Full custom dimensions, explorations, BigQuery, training and governance: 4-7 weeks depending on complexity and number of properties.

Tired of GA4 reports that raise more questions than they answer?

Get a free GA4 audit. We will show you exactly what a properly configured property with real decision data would look like.

Get My Free GA4 Audit arrow_forward