Well, client-side "on-site" trackers are used because it's easy and flexible to do so using JavaScript tags (and then get blocked by privacy aware user agents). But in my experience at a publisher many years ago, most of the important client side measurements tracked (think page views, conversions) ultimately correspond to a browser request made to a server in the publisher's control, such as fetching the article's contents or finalizing a purchase.

It logically follows that the same measurements could have been made by tracking events on the server, possibly without involving additional client side scripts at all. No, it wouldn't be as easy or allow tracking everything the user does down to the pixel, but that's not the adblock enlightened user's problem...

It is my understanding the industry is moving towards server side "tagging" to improve performance and probably also to try and obfuscate it from ad blockers. But I haven't paid much attention to that.

> most of the important client side measurements tracked (think page views, conversions) ultimately correspond to a browser request made to a server in the publisher's control, such as fetching the article's contents or finalizing a purchase.

That won't tell you how many didn't see the page or converted though.