This means that although an outgoing HTTP request to Google Analytics comes from a Google Cloud virtual machine, Google Analytics updates the usage data to refer to the actual device and IP address of the user. However, the Universal Analytics label on the server container copies the IP address of the user and agent to the outgoing protocol request using the &uip and &uip settings. This is sufficient to avoid overwriting, and can ensure that the actual user address and the chain of agents are not transferred to Google Analytics in any form or size. In other words, if you use a server container as a proxy, it “hides” the real user because all the answers seem to come from a virtual machine, not from the user’s browser. Full documentation on the available fields is still missing, but if you check the debug mode of the Server Container and look at the Event Data tab, you will go far. Leaving the fields blank, the ticker is essentially unable to fill these fields with the values of the request received, i.e. the user’s browser and the actual device. This should at least support the idea that the tags on the Google Tag Manager server side can be used to improve the perspective of anonymizing and obscure the user’s personal information. Since the launch of server tags in Google Tag Manager, I have used every opportunity to mark the tools it provides to improve the privacy and security of end-user data. Yes, the search table variable will make things much simpler since all you need is a tag and a trigger and the user sets ip_override and user_agent to empty values or default values depending on whether the IP should be anonymous or not. Another tag sends data to Google Analytics following the instructions in this article. In this article, I will show you how to bypass this behavior and prevent Universal Analytics from performing copy and paste operations. On the Universal Analytics tag that is activated in the server container, you need to perform the following steps.