How do affiliate networks work, at a technical level? I am an experienced web developer, so no need to simplify.
My current project has required me to do some research in the world of affiliate networks. I'm trying to figure out exactly how the existing ones are implemented, names such as rewardStyle, Rakuten, CJ Affiliate, Skimlinks, etc.
First, the publisher side. They link to merchant websites with special links that contain a code that identifies the publisher. I'm going to call this "referral code" (I'm not sure what the lingo is). Apart from that, it seems straightforward enough.
On the merchant side, we need to detect the referral code and notify the affiliate network of a successful conversion. The way I see it there are, theoretically, two ways to do this:
The merchant handles this with backend code. This involves significant development work for every single merchant, which I find unlikely they'll agree to do, but is resistant against anti-tracking measures in browsers.
The merchant uses a tracking script (JS running on the browser), possibly managed with Google Tag Manager or similar. More scalable, but I assume affected by anti-tracking measures?
I can see big names such as Amazon using #1 (and in fact I understand they do), because they are a one single site and not a network. For affiliate networks, I assume everybody does #2.
- Am I correct?
- Are there other techniques I may be missing?
- Are affiliate trackers affected by ad blockers and similar?