iOS 14+ changed everything about Meta Ads measurement. Apple's App Tracking Transparency (ATT) framework limited Facebook's ability to track user behavior across apps and websites.
Here's what you need to know to track Meta conversions accurately in 2026.
What Changed with iOS 14+
Before iOS 14
Meta could:
- Track users across websites and apps
- Build comprehensive user profiles
- Match conversions back to ad views/clicks with high accuracy
- Report near-real-time conversion data
After iOS 14
Meta now has:
- Limited cross-app tracking (users must opt in)
- Reduced event data from iOS devices
- Delayed reporting (up to 72 hours)
- Modeled/estimated conversions
- Maximum 8 conversion events per domain
The impact: Platform-reported conversions dropped 20-40% for many advertisers, even when actual sales didn't change.
The Two Types of Tracking Loss
1. Measurement Loss
This is artificial. Sales still happen—Meta just can't see them.
Symptoms:
- ROAS drops but revenue stays stable
- Blended metrics unchanged while Meta metrics decline
- Gap between Meta-reported and actual conversions grows
2. Performance Loss
This is real. Algorithm optimization suffers with less data.
Symptoms:
- CPA increases across all metrics (not just Meta reporting)
- Blended ROAS actually declines
- Campaigns have trouble exiting learning phase
Most advertisers experience both. The key is distinguishing between them.
Essential Setup: Conversions API (CAPI)
The Conversions API sends event data directly from your server to Meta—bypassing browser restrictions.
Why CAPI Matters
- Works even when cookies are blocked
- Unaffected by ad blockers
- More reliable than pixel-only tracking
- Improves event match quality
- Required for optimal delivery
How to Set Up CAPI
For Shopify:
- Meta Sales Channel integration (automatic CAPI)
- Or use partner apps (Elevar, Triple Whale)
For WooCommerce:
- PixelYourSite Pro plugin
- Or custom integration via Facebook Marketing API
For Custom Sites:
- Implement via Facebook Marketing API
- Send server-side events with user data (email, phone, IP)
CAPI Best Practices
Send both pixel and CAPI events: Meta deduplicates them, but redundancy ensures coverage.
Include user data: Hashed email and phone improve match rates from ~50% to 80%+.
Set up on all key events: At minimum: PageView, ViewContent, AddToCart, InitiateCheckout, Purchase.
Event Configuration: The 8-Event Limit
iOS 14 limited domains to 8 prioritized conversion events.
Recommended Priority Order (Ecommerce)
- Purchase
- InitiateCheckout
- AddToCart
- ViewContent
- AddPaymentInfo
- Lead (if applicable)
- Search
- CompleteRegistration
Purchase must be #1. If a user triggers Purchase and AddToCart on the same session, only the highest-priority event is reported.
Aggregated Event Measurement (AEM)
AEM is how Meta handles iOS 14 events:
- 8 events max per domain
- 72-hour reporting delay possible
- Modeled conversions for opted-out users
- Requires domain verification
Domain Verification
You must verify your domain to:
- Access AEM
- Configure conversion events
- Ensure proper attribution
How to verify:
- Go to Meta Business Suite → Brand Safety → Domains
- Add your domain
- Verify via DNS TXT record (recommended) or meta tag
Improving Event Match Quality (EMQ)
Event Match Quality measures how well Meta can match events to users. Higher EMQ = better optimization.
Target: 6.0+ (Good), 8.0+ (Great)
How to Improve EMQ
Send more user data:
- Hashed email (most important)
- Hashed phone number
- First name, last name
- City, state, zip code
- IP address
- User agent
- Click ID (fbc parameter)
- Browser ID (fbp parameter)
Ensure proper deduplication:
- Send matching event_id for pixel and CAPI events
- Use unique transaction IDs for purchases
Attribution Settings
Attribution Windows
Meta now defaults to:
- 7-day click-through
- 1-day view-through
You can compare in reporting:
- 1-day click
- 7-day click
- 28-day click (limited data post-iOS 14)
- 1-day view
Recommendation: Use 7-day click for optimization, but understand this captures some conversions that would have happened anyway.
Comparing Attribution Settings
Run reports with different windows to understand behavior:
| Window | Typical Reported ROAS | Reality |
|---|---|---|
| 1-day click | 1.5-2.5x | Most conservative |
| 7-day click | 2.5-4.0x | Standard, some inflation |
| 7-day click + 1-day view | 3.0-5.0x+ | Most inflated |
The gap between 1-day click and 7-day click reveals how much credit Meta takes for delayed conversions.
Third-Party Solutions
Platform tracking will never be complete. Many brands supplement with:
Triple Whale
- Multi-touch attribution
- First-party tracking pixel
- Blended ROAS calculations
- $79-349/month
Northbeam
- Incrementality modeling
- Customer journey tracking
- Budget optimization
- Enterprise pricing
Elevar
- Server-side tracking
- Data layer management
- Privacy-compliant
- $150-500/month
Rockerbox
- Multi-touch attribution
- TV/streaming integration
- Enterprise focus
These tools help, but none solve the problem completely. Plan to operate with some uncertainty.
Practical Measurement Framework
Given tracking limitations, here's how to actually measure Meta performance:
1. Track Blended Metrics
MER (Marketing Efficiency Ratio):
MER = Total Revenue / Total Ad Spend
If MER stays stable when you increase Meta spend, Meta is working—even if platform ROAS looks low.
2. Run Holdout Tests
Turn off Meta ads for a region or audience segment. Measure the revenue impact over 2-4 weeks. This reveals true incrementality.
3. Look at Leading Indicators
- New customer acquisition rate
- First-time purchaser volume
- Traffic from direct/organic (if down when ads are off, ads were working)
4. Trust Directional Trends
If Meta ROAS shows +20% after creative tests, actual performance likely improved—even if absolute numbers are wrong.
Common Mistakes
1. Panicking Over Reported ROAS Drops
If your blended metrics are stable, the issue is measurement—not performance.
2. Not Setting Up CAPI
Running Meta Ads without CAPI in 2026 means accepting 30-40% data loss. It's mandatory.
3. Ignoring Event Match Quality
Low EMQ (under 5.0) means your tracking is leaking data. Send more user data to Meta.
4. Over-relying on Platform Reporting
Build a measurement framework that doesn't depend entirely on what Meta tells you.
The Bottom Line
iOS 14 didn't break Meta Ads—it broke Meta Ads measurement.
To navigate this:
- Implement CAPI properly
- Maximize Event Match Quality
- Verify your domain
- Use blended metrics alongside platform reporting
- Run incrementality tests to validate
- Accept some uncertainty as permanent
The advertisers who succeed post-iOS 14 are those who built robust measurement systems—not those who waited for Meta to fix everything.