Attribution Model is the time that the user has to convert and how this is calculated on the platforms. Both Facebook and Google Analytics account for this conversion in different ways. See the detailed explanation below to understand it a little better.
Facebook uses by default the attribution model based on 1 day of viewing and 28 days of interaction. In other words, if someone clicked on your ad in a 28-day window and converted, or viewed your ad and converted in one day, the conversion is attributed to Facebook.
Google Analytics, on the other hand, uses the non-direct last click model, that is, it attributes 100% of the conversion value to the last channel with which the user interacted, before completing the conversion.
Since Analytics only considers interaction (i. e. clicks) with the ad, it will never consider Facebook ad views. And it's also important to take into account that the user tends to interact in different ways with the sponsored links before actually converting. The user’s purchase path is increasingly complex, interacting on more than one device and with more than one channel. This means the results on Google Analytics will rarely match those observed in ads platforms.
Every platform uses the attribution model that's most beneficial to itself. This way, results may overlap, meaning that one conversion can be attributed by more than one platform by itself. For this reason, Analytics is important to track the overall business result, not just a single channel. But, for our part, we only use it for this purpose: to monitor the general result. On each individual platform, we can see the performance of each one separately.