Set up properly, not just installed
GA4 implementation with the event structure and custom dimensions your business actually needs.
Where this usually breaks down
A GA4 property with the default setup looks complete even when it is not. Enhanced measurement fires automatically, a handful of events show up in reports, and it is easy to assume the platform is working. What it is usually missing is anything tied to how the business actually operates - custom conversion events, dimensions that map to real segments, and e-commerce data that reconciles against actual orders.
The result is a property that reports activity without reporting meaning. Traffic and engagement numbers look fine, but nobody can answer a specific question like which lead source produces customers who actually renew, because the dimension needed to answer it was never configured.
What this service actually solves
A proper GA4 implementation starts from the business questions that need answering, not from the list of events GA4 collects by default. Custom events, dimensions, and metrics are built to capture the actions that actually matter - a demo request, a return customer, a specific product category - and e-commerce tracking is validated against the systems that process real transactions, not assumed to be correct because it fires.
GA4 implementation is configuring events, dimensions, and e-commerce tracking to match your specific business model, not just installing the default tag.
How we run it
We start by mapping the questions the business actually needs the data to answer, then design the event and dimension architecture backwards from those questions. Every implementation is validated in debug view before it goes live, and e-commerce and conversion numbers are checked against order or CRM data rather than trusted on faith - a tag that fires is not the same thing as data that is correct.
Capabilities & deliverables
Event & Conversion Architecture
- Custom event design mapped to real business actions
- Conversion event configuration and threshold tuning
- Micro vs macro conversion structuring
Custom Dimensions & Metrics
- Correct scoping across user, event, and item level
- Dimension definitions tied to actual reporting questions
Cross-Domain & Cross-Platform Tracking
- Cross-domain measurement configuration
- App and web property linking with a consistent user ID
E-Commerce & Enhanced Measurement
- Enhanced measurement configuration and cleanup
- Purchase funnel and item-scoped event tracking
- Refund and return event handling
Data Validation
- Validation against CRM or order-management data
- Debug view QA before anything goes live
How an engagement runs
Discovery & Question Mapping
We document the specific business questions the property needs to answer before touching any configuration.
Event & Dimension Architecture
Events, dimensions, and metrics are designed backwards from those questions, not from the default event list.
Implementation
Configuration is built through GTM or gtag, whichever the existing setup calls for.
Debug View QA
Every event is checked in debug view before it is allowed to go live.
Validation Against Source Systems
E-commerce and conversion numbers are checked against CRM or order data, not assumed correct because they fire.
Handover & Documentation
The final event and dimension map is documented so it survives beyond whoever built it.
How this compares
| Configured GA4 Property | Default Installation |
|---|---|
| Events map to real business actions | Only automatically collected events are captured |
| Custom dimensions answer specific reporting questions | Generic dimensions with no business context |
| Numbers are validated against source systems | No verification that the data is accurate |
The tag firing is not the same thing as the data being usable - that gap is exactly what a proper implementation closes.
What this changes for the business
- Reports reflect actual business actions rather than generic pageviews
- Custom dimensions answer the questions stakeholders actually ask
- E-commerce numbers reconcile against order data instead of diverging silently
- Cross-domain journeys are tracked as a single user, not counted twice
Who needs this
Businesses relying on the default GA4 setup
If nobody configured events beyond what GA4 auto-collects, reporting is missing the actions that actually matter to the business.
Sites with multi-domain or app-and-web journeys
Without cross-domain and cross-platform linking, the same visitor gets counted as multiple people.
Related work
We're still building out published proof for this specific service — ask us directly and we'll walk through relevant examples.
Common questions
Turning it on gives you enhanced measurement and a handful of automatic events. It does not give you conversion events, dimensions, or e-commerce tracking tied to your actual business model, which is where most of the useful reporting comes from.
Yes - custom platforms just mean the data layer events have to be built manually rather than pulled from a pre-built plugin. The underlying GA4 configuration is the same either way.
In most cases we work with the existing container and correct or extend it. A full rebuild is only recommended when the existing setup is broken enough that fixing it piece by piece would take longer.
E-commerce and conversion events are checked against order or CRM records during implementation, and again after launch once real traffic is flowing through the property.
No - some divergence is normal due to ad blockers, consent choices, and timing differences between systems, and any agency claiming an exact match either is not looking closely or is not being straight with you. What we deliver is a documented, explainable gap instead of an unexplained one.
UA goals cannot be migrated directly since the platforms are structurally different, but we rebuild the equivalent conversion logic in GA4 based on what those goals were actually measuring.
Not sure your GA4 events reflect real business actions?
We will check what is actually being tracked against what your business actually needs to know before recommending any changes.
Talk to us →