CAPI is short for Conversion API and is the latest method for tracking conversions with Facebook. The Conversions API is used for tracking users, events and conversions to supplement the tracking done by the Meta Pixel (previously called the Facebook Pixel).
Meta introduced the Conversions API in 2021 as a means of addressing the increasing user privacy protections being rolled out by various platforms, such as iOS, and browsers over the last few years. Unlike the Pixel, which is a browser based tracking system, CAPI utilises a server-to-server method for sending events to Facebook/Meta, allowing it to circumvent many of the blockers that now exist and is in line with the general trend we’re seeing in the digital marketing space.
To understand how CAPI works, you first have to understand how the Pixel works. When using the Pixel, you typically install it on the “front-end” of your website (i.e the part of your website that loads within a user’s browser). When a user then loads up your website, the Pixel is loaded within their browser and sends data to Facebook from the user’s browser/device. This makes it susceptible to blockers from the user’s end, whether it be an ad/tracking blocker installed on their browser or iOS14’s data consent settings.
CAPI attempts to avoid these blockers by using a server-to-server setup. As opposed to a user’s browser/device sending the data to Facebook, a web server is used instead. This means that platforms and companies can have greater control, but also send more accurate data to Facebook.
Though there are downsides to CAPI. Firstly, they require access to a web server. If you have built your website on a platform, such as Shopify, then you likely don’t have access to those servers and are reliant on those platforms to implement CAPI themselves. Secondly, setting up CAPI requires developer resources. It is not as simple as adding a snippet of code onto a web server, like it is with the Pixel. CAPI requires planning and code to be written specifically to implement it and any subsequent changes will also likely require code to be changed or more to be written.
CAPI isn’t meant to replace the Meta Pixel, rather it’s meant to supplement it. By running both a CAPI implementation as well as a Pixel at the same time, you can ensure the data you’re sending to Facebook is both quality and accurate. This is done through Event Matching.
Event Matching refers to when Facebook is able to match the events sent by both the Pixel and CAPI together. An example of this is when a user purchases something off of a store, the event is sent via both the Pixel and CAPI. Then through various pieces of data, such as an order ID, Facebook is able to see that both events come from the same user’s purchase and can treat that as quality data.
To do this though, Facebook requires various pieces of information to be sent with these events, such as:
Email Address (in a hashed format)
But when looking at how website’s work realistically, not all of these pieces of information are available all the time. For example, a website may not know a user’s email address when they first visit a store and start adding products to their cart. So with events such as Add To Cart, it may not be possible to have a high event match rate. You should keep this in mind when looking at Facebook’s Events Manager and seeing low match rates.
CAPI is Meta’s current answer to the increasing user privacy issues that are currently trending in the online space and is indicative of the direction that user tracking is heading toward in the near future. While server-to-server tracking implementations can provide accurate and complete information on users, it also requires more strategic planning and developer work to get right.
Looking to start using GA4?