Problem diagnosis · Meta Ads
If you're running Meta Ads with only a browser pixel, you're missing 25–40% of purchase events before they ever reach Meta's algorithm. Your ROAS looks fine. Your actual revenue doesn't match. The fix is the Conversions API — but only if it's set up correctly.
I'll review your Meta Events Manager, pixel setup, and CAPI configuration. You'll get a written summary of what's broken and what your EMQ score should be — within 24 hours.
Get a free CAPI audit →+22% match rate — typical improvement
When Apple introduced App Tracking Transparency in 2021, most users opted out. On iOS, the Meta Pixel can no longer reliably track purchases — click IDs get stripped from URLs, attribution windows shrink, and the pixel sees fewer conversions with less identifying information. For e-commerce stores with high mobile traffic, this alone can cut reported conversions by 30–40%.
uBlock Origin, Brave, and Firefox Enhanced Tracking Protection block Meta's pixel domain outright. In the Netherlands and Germany, ad blocker adoption exceeds 30% of desktop users. Every one of those visitors is invisible to your pixel — including when they buy.
You installed CAPI — but now Meta is counting the same purchase twice: once from the browser pixel, once from the server. Your ROAS looks inflated. Your CPA looks artificially low. Meta's algorithm is optimising toward phantom conversions. This happens when deduplication event_id parameters aren't set correctly.
EMQ measures how effectively Meta can match your conversion events to actual Meta user accounts. A score below 6.0 means Meta is receiving events but can't connect them to users reliably — so Smart Bidding optimises on incomplete audience data. EMQ below 5 is essentially the same as no CAPI.
Most merchants think enabling CAPI in the Shopify Facebook & Instagram channel gives them server-side tracking. It doesn't. Shopify's default integration still requires the browser's Web Pixel to fire in order to trigger the server event. If the browser is blocked, the server event never fires either. This is why stores with "Maximum" data sharing still see EMQ scores stuck at 4.0–6.0.
| Scenario | iOS 14+ users tracked | Ad blocker users tracked | Cookie lifetime | EMQ score | Deduplication |
|---|---|---|---|---|---|
| Browser pixel only | ~20–25% | 0% | 7 days (Safari) | 3–5 | N/A |
| Shopify native CAPI | ~50–60% | ~30% | 7 days (Safari) | 4–6 | Inconsistent |
| CAPI via sGTM + Stape | 90–95% | 90%+ | 400 days | 7–9+ | Reliable |
A misconfigured CAPI duplicates events and corrupts your audience data. I'll check yours and tell you exactly what's happening — no cost, no commitment.
Get a free CAPI audit