Why don't our numbers match?
Reporting numbers never match across platforms because each one measures differently: Meta counts view-through conversions in its attribution window, GA4 typically uses data-driven or last-click models, and consent banners block 20–40% of tracking entirely in privacy-strict regions. The platforms will never reconcile to 100%, the correct response is to designate one source of truth per KPI rather than chase perfect agreement.
Key facts
- Meta attributes on views and clicks within its window; GA4 uses different models, the same sale legitimately appears differently in each, or in both.
- Cookie-consent rejection removes an estimated 20–40% of client-side tracking in regions with strict enforcement.
- Platform-vs-analytics discrepancies of 20–50% are normal; the fix is a defined source of truth per KPI, not reconciliation of every click.
The four discrepancy sources
Attribution windows (7-day click vs. 30-day vs. view-through), tracking method (client-side pixels vs. server-side events vs. CRM records), consent and browser blocking (ITP, ad blockers, consent rejection), and time-zone/date-boundary differences. Each alone creates gaps; together they guarantee platforms disagree.
What this means for decisions
Trying to reconcile every click wastes analyst time on an impossible task. Instead, define the source of truth per decision: platform metrics for in-platform optimization, analytics for cross-channel comparison, CRM/transaction data for business reporting. Each number is 'right' for its job.
How to shrink the gaps
Standardize UTM parameters ruthlessly (one taxonomy, enforced at trafficking), implement server-side tracking (Meta CAPI, GA4 Measurement Protocol) to recover browser-blocked events, and align attribution windows where platforms allow it. Teams typically recover a meaningful share of 'missing' conversions with server-side implementation alone.
When to run an attribution audit
Bring in external support when discrepancies exceed ~50%, when finance stops trusting marketing numbers, or before any major budget reallocation, an attribution audit determines which data sources you can actually trust for which decisions.
Frequently asked questions
Why do Meta and GA4 report different conversion numbers?
They measure differently: Meta claims view-through and click conversions inside its attribution window; GA4 applies its own model. Consent blocking and browser privacy remove another 20–40% of client-side events. They will never match exactly.
Should we try to reconcile platform numbers?
Not click-by-click. Define one source of truth per KPI, platform data for in-platform optimization, analytics for cross-channel views, CRM data for revenue reporting, and use each for its purpose.
How do we reduce tracking discrepancies?
Enforce one UTM taxonomy, implement server-side tracking (CAPI / Measurement Protocol) to recover browser-blocked events, and align attribution windows across platforms where possible.