Basically it’s client side vs server side, and how privacy regulations are making it harder to do it the old way. Pixels/cookies are saved to your browser (client side) while postback tracking communicates the action without your involvement (server side).
The growing use of multiple browsers and devices made the client side tracking quite difficult because each instance is considered unique even though it’s by one person, so the industry got around this by sharing that cookie data between browsers/devices which then created privacy concerns. Sure, everyone talked about privacy but no one really cared enough to change the practice until Apple released iOS14 a little over two years ago.
Server-to-server (S2S) tracking, on the other hand, doesn’t rely on your browser or cookies. Instead, it involves communication between the servers of the advertiser and the source (publisher). When an action is completed, like you clicking an ad or making a purchase, the publisher’s server sends a notification (postback) to the advertiser’s server. This means the tracking is done behind the scenes, without involving your browser or storing files on it.
The problem is this inherently gives less events to track off of, which makes lead/sales flow less accurate. So advertisers are embedding unique “click IDs” into url parameters (gclid or fbclid for example) to add fidelity to the process. But, as with everything in this industry, this practice will start to go away soon as Apple is releasing iOS17 which is starting the process of eliminating click ids between apps. You don’t need a crystal ball to see that this trend will only continue and gclids/fbclids will eventually go away.
It’s hard to both ELI5 and explain in great detail, so I hope I covered enough to make sense of it all.