Measurement is the least interesting part of advertising and the one that decides everything. The algorithm optimizes toward whatever you tell it success looks like, so a broken signal produces confidently wrong spending.
Two ways events reach Meta
- The pixel: browser-side tracking. Easy to install, increasingly lossy thanks to blockers, privacy settings, and iOS changes.
- The Conversions API (CAPI): server-side events sent from your backend or platform integration. More reliable, and now essentially table stakes rather than an advanced option.
Most platforms — Shopify, WooCommerce, common site builders — have a supported integration for both. Use both, with event deduplication configured, rather than choosing one.
Optimize for the real thing
Tell the platform to optimize for the event that is actually worth money — purchase, or a qualified lead — not for clicks, landing page views, or "engagement." If your volume is too low for the platform to learn on purchases, optimize for the closest upstream event that still correlates with revenue, and check the correlation rather than assuming it.
Attribution will not match your books
In-platform numbers, your analytics, and your bank account will all disagree, sometimes wildly. Reasons include attribution windows, view-through credit, cookie loss, and cross-device journeys.
Practical response: use platform data for relative decisions — which ad, which audience, which creative — and use your own revenue for absolute decisions about whether the whole channel is working. Blended CAC, meaning total spend divided by total new customers, is the number that cannot lie to you.
Try it
Make a test purchase or submit a real lead and confirm the event appears with the right value. Then calculate blended CAC for last month from your own books, and keep tracking that alongside whatever the platform reports.