Apple's App Tracking Transparency framework permanently changed how a meaningful share of Meta ad performance gets measured, and advertisers who haven't adapted their measurement setup accordingly are often making decisions on incomplete or degraded data without realizing it. Getting this right is foundational, not optional, at any serious spend level.
What actually changed
Since iOS 14.5, users must opt in to app tracking, and a substantial share choose not to — which limits the deterministic, user-level event data available for those users. Meta's response is Aggregated Event Measurement, a framework that reports and optimizes against a prioritized, limited set of conversion events per domain using modeled and aggregated data rather than clean individual-level tracking for opted-out users.
This doesn't mean iOS measurement is broken or unusable — it means it's structurally different, relying more heavily on statistical modeling to fill gaps, and advertisers need to understand and work within that reality rather than expecting pre-2021 measurement precision.
Conversions API as the essential countermeasure
Server-side event tracking via the Conversions API sends conversion data directly from your server to Meta, independent of browser-based tracking limitations, and — when properly deduplicated against your browser pixel — meaningfully recovers signal that would otherwise be lost. This has moved from an advanced technical option to baseline required infrastructure for any advertiser serious about measurement accuracy.
Implementation quality matters as much as implementation itself — a poorly configured CAPI setup with incomplete event parameters or broken deduplication logic can actually create double-counting or inconsistent reporting, so verify event match quality and deduplication status directly in Events Manager rather than assuming a CAPI integration is working correctly just because it's installed.
- Implement CAPI as baseline infrastructure, not an advanced optional step
- Verify event match quality and deduplication in Events Manager, don't assume it's correct
- Prioritize conversion events deliberately through Aggregated Event Measurement
- Reconcile platform-reported conversions against total business-level revenue regularly
Setting realistic expectations for reported numbers
Even with strong CAPI implementation, expect some genuine gap between platform-reported conversions and total actual conversions, particularly for iOS traffic — this is a structural feature of the current measurement environment, not a sign your setup is broken. The practical response is to treat platform-reported ROAS and CPA as directionally reliable for optimization and relative comparison, while periodically reconciling against your actual business-level revenue and order data for the true picture.
Blended metrics — total spend across all channels against total revenue, independent of any single platform's attribution — remain one of the most reliable sanity checks available, precisely because they don't depend on any one platform's measurement model.
Event prioritization as an active decision
Within Aggregated Event Measurement, you actively rank which conversion events matter most for a given domain (typically purchase above add-to-cart above view-content) — this ranking directly determines what gets measured and optimized against for users where full tracking isn't available. Review and update this prioritization deliberately as your business priorities shift, rather than leaving it at whatever was set during initial domain verification.
This connects directly to the domain verification requirements covered in the Account Infrastructure category — a domain that isn't verified can't have its event prioritization configured at all, which is a common and entirely avoidable measurement gap.
Building measurement resilience as spend scales
The higher your spend, the more a measurement gap costs you in misallocated budget, which is why Power Ads treats measurement infrastructure — CAPI, domain verification, event prioritization — as a first-order onboarding priority for scaling clients, not a secondary technical detail to address later. Solid measurement is what makes every other optimization decision in this series (bidding, targeting, creative testing) actually trustworthy.
Key takeaways
- Aggregated Event Measurement relies on modeled data for opted-out iOS users, not clean deterministic tracking
- CAPI is baseline required infrastructure, and implementation quality matters as much as having it
- Expect and plan for a genuine measurement gap; reconcile against blended business-level revenue regularly
- Event prioritization in Aggregated Event Measurement is an active decision, not a set-and-forget default
FAQ
Does CAPI fully solve the post-iOS measurement gap?
It meaningfully recovers lost signal but doesn't fully close the gap for opted-out users — expect some genuine, structural difference between platform-reported and total actual conversions.
How often should I update my event prioritization?
Review it whenever your business priorities or funnel shift meaningfully, and at minimum revisit it during any major account or campaign restructuring.
