The assumption
At some point, the conversation shifts.
Your data doesn’t match reality.
The same metric shows different values in different reports.
Teams stop trusting the numbers.
And the conclusion becomes:
“We’ve outgrown GA4. We need GA360.”
It sounds reasonable.
More power.
More scale.
More reliability.
But this is where the decision usually goes wrong.
GA360 typically starts around $50,000 per year—and can increase significantly with scale.
This isn’t just a tooling decision.
It’s a meaningful investment.
So the real question becomes:
Are you upgrading your system—or just your costs?
What people think GA360 solves
When companies consider GA360, they’re typically trying to solve:
- data discrepancies
- missing or unreliable tracking
- inconsistent reporting
- lack of trust in dashboards
- limitations in analysis
The belief is simple:
Better tooling will produce better data.
What actually happens
GA360 doesn’t fix broken data.
It processes it more efficiently.
If your system is misaligned:
- events are still misdefined
- identity is still inconsistent
- logic is still fragmented
- meaning is still unclear
You just get:
faster access to unreliable data
This is why teams upgrade—and still don’t trust their numbers.
This is not a tooling problem
Analytics platforms sit at the end of the pipeline.
They display and aggregate data.
They do not define:
- what data is collected
- how it’s structured
- how it connects across systems
- what it actually means
Those decisions happen upstream.
This is where most systems break—within the structure of the data estate itself (see: What is a Data Estate).
Where the system actually fails
When companies experience data issues, the root cause is usually structural:
1. Collection layer issues
- incomplete or inconsistent tracking
- weak or missing data layer
- reliance on page-level logic instead of system design
This is where the signal first breaks—especially at the collection layer (see: What Good Event Tracking Actually Looks Like).
2. Processing and modeling gaps
- no consistent transformation layer
- business logic scattered across tools
- no defined schema
This creates fragmentation—especially when business logic isn’t centralized in a defined layer (see: Where Logic Belongs in a Data Estate).
3. Missing memory layer
- reliance on GA4 as the system of record
- no structured warehouse (e.g. BigQuery) to store and control historical data
- no control over how data evolves over time
This limits what the system can become—especially without a persistent data layer to store and work with historical data (see: How GA4 BigQuery Export Changes Everything).
4. No semantic layer
- inconsistent definitions across teams
- unclear metrics
- conflicting interpretations
This is where trust breaks—when there’s no shared layer defining metrics and meaning across the system (see: Semantic Layer).
A simple example
If a “purchase” event fires inconsistently across different checkout flows:
- some transactions are tracked
- others are missed
- some are duplicated
GA360 won’t resolve that inconsistency.
It will just report it more reliably.
Why GA360 feels like the answer
GA360 is often introduced at the moment when:
- data complexity increases
- reporting demands grow
- teams need more flexibility
So it appears to be the next logical step.
But what’s actually happening is:
The system is under strain—and the structure hasn’t kept up.
GA360 doesn’t resolve that strain.
It exposes it.
The real implication
Upgrading tools without fixing structure creates a more complex version of the same problem:
- more data, same inconsistencies
- more reporting, same confusion
- more cost (often $50K+ per year), without resolving the underlying issue
This is why:
Teams can spend significantly more—and still not trust their data.
When GA360 actually makes sense
GA360 can be the right decision when:
- data volume exceeds standard GA4 limits
- sampling impacts critical reporting
- service-level agreements (SLAs) matter
- enterprise-level support is required
In these cases, GA360 is a scaling decision—not a fix for underlying data issues.
A note on rollup reporting
One common reason companies consider GA360 is the ability to create rollup properties—combining data across multiple sites or properties into a single view.
This is a valid requirement.
But it’s often misunderstood.
Rollup reporting is not inherently a GA360 capability—it’s a data modeling capability.
With a structured data pipeline (e.g. using BigQuery and a defined transformation layer), it’s possible to:
- combine multiple data sources
- standardize schemas across properties
- create unified reporting views
In other words:
You don’t need GA360 to create rollups—you need a system that supports them.
Before you upgrade
Before moving to GA360, ask:
- Is our tracking consistent and complete?
- Do we have a defined data structure?
- Are our metrics clearly defined and agreed upon?
- Can we explain discrepancies today?
If the answer to these is no:
GA360 won’t solve the problem—it will just make it more visible.
What actually changes when the system is fixed
When the system is structured correctly:
- tracking becomes consistent
- data flows predictably
- definitions are stable
- analysis becomes reliable
At that point:
GA4 is often sufficient.
GA360 becomes a scaling decision—not a correction.
Connection to the broader system
This decision sits inside a larger pattern:
- treating symptoms at the interface layer
- ignoring structure upstream
- expecting tools to resolve system issues
This is the same pattern behind unreliable dashboards, broken attribution, and untrustworthy AI outputs—problems that originate upstream but surface at the interface layer (see: Why Analytics Data is Wrong).
Final takeaway
If your data doesn’t make sense:
Don’t start with a platform upgrade.
Don’t start with reports.
Start with the system.
Because:
GA360 won’t fix broken data.
It will just help you see it more clearly—at a higher cost.
Where to go next
If you’re considering GA360 because your data isn’t reliable:
Before you commit to GA360, validate your system.
See Evaluate.
Doug McCaffrey
Designs and maintains analytics systems that remain reliable over time.
Explore how this connects across your data estate: