Data Discrepancy Between GA4 and AMP

Why don’t GA4 and AMP data match exactly?
1.1 Google Data Processing Window
1.2 Session Deduplication
1.3 The role of utm_id
Recommended Metrics for Lower Variance
Immediate Actions to Minimize Differences
Frequently Asked Questions
Additional references

Why don’t GA4 and AMP data match exactly?

It is common to observe variations between Google Analytics 4 and Adsmurai Marketing Platform. This is not a measurement error, both tools are built on different architectures and designed for different purposes.

Understand the 3 main reasons behind these differences and how to minimize their impact on your reporting.

1. Google Data Processing Window (up to 30 days)

Google Analytics 4 does not finalize data immediately. It can take up to 30 days to fully consolidate conversions, attribution, and sessions.

Timeframe	Data Status	Reliability
Today	Provisional	⚠️ Low
7 days	Partially consolidated	✅ Medium
30 days	Final	✅ High

How AMP handles this:

We retrieve daily “snapshots” to ensure fast data visualization
We automatically refresh data from 1, 2, 7, 15, and 30 days ago
Each refresh incorporates Google’s latest adjustments

Practical impact:

If you check “yesterday’s” data today and then review it again next week, it will likely change slightly. This is expected and reflects Google’s data consolidation process.

2. Session De-duplication: Unique Users vs Historical Aggregation

This is one of the most common causes of discrepancies between GA4 and AMP.

How each platform works:

Aspect	GA4 (Interface)	AMP
Data source	Real-time data with user context	Pre-aggregated historical data
When breaking down by dimension	De-duplicates sessions by unique user	Sums all available sessions

Example:

User A → 1 session in GA4 (even if they had 3 visits)
User A → 3 sessions in AMP (direct sum of visits)

What this means in practice:

Real scenario:
A user visits your website 3 times in one day from the same Facebook campaign.

GA4: Filtering by “Facebook Campaign” shows 1 session (user-level de-duplication)
AMP: Shows 3 sessions (sum of all visits)

Why does this happen?

GA4 has access to unique user identifiers in real time and applies de-duplication dynamically
AMP works with historical, dimension-aggregated data, meaning it sums sessions without user-level context

Is this an issue?

No, both perspectives are valid:

GA4: Best for understanding reach and unique users
AMP: Better for measuring traffic volume and cumulative campaign impact

3. Campaign Matching: The Role of `utm_id`

To match a GA4 session with a specific paid media campaign, AMP requires a reliable unique identifier.

Mandatory requirement:

utm_id={campaign_id}

Full example for Meta (Facebook/Instagram):

utm_id=_caid_adsetid_adid

This structure includes:

campaign.id: Campaign ID
adset.id: Ad set ID
ad.id: Ad ID

Common scenarios where matching fails:

Scenario	Result in AMP
utm_id not configured	⚠️ Session not attributed
Incorrect campaign ID	⚠️ Match failure
Campaign paused (no spend that day)	⚠️ Not shown in Paid Media reports
Ad format without dynamic fields	⚠️ ID not available

What can you do?

Campaign attribution accuracy depends directly on correct utm_id configuration. Make sure that:

✅ All new campaigns include a utm_id
✅ The format follows the recommended structure
✅ Dynamic parameters (, etc.) are correctly implemented in the ad platform
✅ Legacy campaigns have been audited and updated if needed

💡 Recommended Metrics for Lower Variance

If you need more consistency between GA4 and AMP, consider alternatives to sessions:

Metric	GA4 vs AMP Variance	Recommended Use Case
Users	⚠️ Lower variance	Audience reach, unique users
New Users	✅ Very low variance	New user acquisition
Engaged Sessions	✅ Very low variance	Traffic quality, user engagement
Sessions	⚠️ Higher variance	Total traffic volume (with caveats)

Why are these metrics more stable?

Users / New Users: Based on unique identifiers, making de-duplication more consistent across platforms
Engaged Sessions: Defined by GA4 using strict criteria (≥10s duration, conversion, or 2+ pageviews), reducing aggregation ambiguity

Recommendation:

For client-facing performance reports where consistency is key, prioritize Engaged Sessions or New Users over total sessions. This reduces perceived discrepancies while maintaining analytical value.

🎯 Immediate Actions to Minimize Differences

Priority	Action	Owner
🔴 High	Ensure all campaigns have utm_id configured using the full format	Marketing Services Team
🔴 High	Audit active campaigns and update utm_id if outdated formats are used	Marketing Services Team
🟡 Medium	Use data with ≥7 days of maturity for finalized reporting	Analyst
🟡 Medium	Prioritize Engaged Sessions or New Users in client reports	Marketing Services Team
🟢 Low	Set expectations with clients (5–15% variance is normal)	Customer Success

❓ Frequently Asked Questions

Is a 5–10% discrepancy a concern?

No. Differences up to 10–15% are expected due to the factors explained above. Variances above 20% should be investigated.

What should I use for client reporting?

Daily optimization → AMP data (more up-to-date, aligned with spend)
Monthly/quarterly reports → Consolidated data (≥7 days old)
User analysis → Users or New Users (higher consistency)
Traffic quality → Engaged Sessions (highly stable across platforms)

How do I identify the freshest data?

In AMP, you can check the latest data extraction timestamp in the data source sync section.

Can I update utm_id formats in active campaigns?

Yes, but historical data will retain the previous format. New data will follow the updated structure. For consistency, we recommend updating all active ad sets.

Which metric should I use for conversion reporting?

It depends on your objective:

E-commerce → Use specific conversion metrics (purchase, add_to_cart) with lower variance
Lead generation → Use custom conversion events configured in GA4
Engagement analysis → Combine Engaged Sessions with conversion events for a complete view

📚 Additional References

UTM configuration guide to ensure GA4 data is correctly reflected in AMP