Definition: client, server container and vendors
Google's introduction to server-side tagging describes it as moving tag code from the web page to a server container running in a cloud environment the business controls. Three parts make up the arrangement:
- The browser, which still runs a small amount of code: the web container or the Google tag, sending events to the business's own endpoint instead of to each vendor.
- The server container, which receives each request, turns it into a standard event, and runs server tags. In Google Tag Manager, incoming requests are claimed by components called clients, and outgoing calls are made by tags, so one incoming event can feed many vendors.
- The vendors, analytics and advertising platforms that now receive their data from the server rather than from the visitor's browser.
The server is usually mapped to a subdomain of the site, so that requests stay on the site's own domain. The containers and workspaces guide explains how a server container sits beside the web container in the same account, with its own versions and permissions.
In practice: control, performance and data minimisation
The first change is control. Every field that reaches a vendor passes through a place the business can inspect and edit, so identifiers and customer details can be removed before they reach vendors that do not need them, and enriched where a vendor does. The second is performance: fewer vendor scripts in the browser and fewer outbound requests from the page. The third is durability: cookies set by the site's own server tend to survive browser restrictions aimed at cookies written by scripts.
Server-side tagging is also the usual route for server-to-server feeds that do not depend on the browser, such as conversion APIs and affiliate postbacks; the guide to how affiliate tracking works describes how browser limits have pushed matching onto servers. The server-side tagging basics guide lists each improvement with its boundary, and the Google Tag Manager hub places it among the other container topics.
What it does not change: consent and accuracy of the source
Two limits are often misunderstood. First, consent still governs. The banner and consent mode run in the browser, where the visitor chooses, and the server should enforce the consent state it receives, not bypass it. A visitor who declined is not available to measure merely because the request now goes to a different domain. Second, the server only knows what the browser tells it. An event that is wrong at the source, fired on the wrong page or missing a value, is forwarded wrong, now from a server.
There are also costs that the browser setup never had: hosting, monitoring, certificates, updates and an endpoint whose failure silences every tag behind it. The client-side versus server-side tagging comparison weighs those costs against the gains and sets out when each setup fits.
Server-side tagging versus a platform's conversions API
The two are often confused because both send events from a server. A conversions API is a single vendor's endpoint for receiving events from any server: a shop platform, a CRM or a custom back end. Server-side tagging is a general architecture that can feed many vendors, and one of its common uses is to call those APIs from a server tag. A business can send Meta events through its API without a server container, and can run a server container that never touches Meta. The Conversions API guide explains the deduplication with the pixel that either route needs.
Eleven tags reduced to one stream
Spindlecrest Cycle Parts ran 11 vendor tags in the browser: analytics, three ad platforms, two affiliate networks, a reviews widget's tracking, a heatmap tool and three others. The product page loaded 1.4 MB of third-party script, and two vendors were receiving customers' email addresses that no one had approved sending. The retailer set up a server container on a subdomain of its site, kept one Google tag in the browser, and moved eight vendors to server tags; three tags that nobody could explain were removed. A transformation stripped email addresses from every outgoing request except the two ad platforms with enhanced matching enabled. Third-party script on the product page fell to 0.4 MB, and the hosting cost about £120 a month.
Related terms
GTMTag management system. provides the server container type and the interface used to run it. A Conversions APIServer-side event sharing with Meta. is one of the most common destinations for server tags. The Data LayerA structured object on a web page that passes values to a tag manager for tags to use. on the page still describes what happened, because the server can only forward what the browser sends. Consent ModeA Google tag feature that adjusts what tags collect according to the visitor's consent choices. sets the consent state in the browser, which the server container should respect for every vendor it calls.
Common confusions
- Server-side tagging gets around consent. It moves where tags run, not what people agreed to.
- It removes all scripts from the page. At least one tag still runs in the browser.
- It fixes bad data. Errors in the events are forwarded faithfully.
- It is the same as a conversions API. An API is one vendor's endpoint; server-side tagging can feed many.
- It is free once set up. Hosting and operations are ongoing costs.
