Do I need the Conversions API if I already have the Meta pixel?

Yes, in almost every case - but it replaces part of what your pixel does rather than sitting alongside it, and setting it up wrongly costs you more than not having it.

The browser pixel runs in the shopper's browser, which is exactly where tracking gets blocked. Safari's tracking prevention, iOS, ad blockers and Facebook's own in-app browser all interfere with it, so a meaningful share of real purchases never reach Meta. The Conversions API sends the same purchase from a server, where none of those apply.

The catch is that if both send the same purchase and Meta cannot tell they are the same event, it counts two. That is not a rare edge case; it is the default outcome unless the two events carry a matching event id.

Meta deduplicates a browser event against a server event when both carry the same event name and the same event id. Not a similar id, and not the same order value or timestamp - the same id.

This is why 'just turn on CAPI as well' is bad advice. Two tools sending the same purchase with their own id formats produce two counted purchases, inflated reported conversions, and a Smart Bidding model optimising against a number that is roughly double reality. The reporting damage is visible; the bidding damage is not.

The order of operations matters more than any individual step, because getting it wrong is what double-counts.

The honest test is not whether Meta's reported conversions went up - they will go up either way, including when you have accidentally double-counted. The test is whether the platform's count moved closer to the number of orders you actually got.

That comparison needs both numbers in one place: what the platform reports, and what your store genuinely sold. Meta cannot do it, because Meta does not know your order count. This is the specific gap Vibel was built to close - it holds the Shopify order ledger and the tracking data together, so coverage is a division rather than an estimate.