Server-side measurement moves the sending of data from the visitor's browser to a company's own server, which then sends it on to analytics and advertising systems. This resolves part of the problems with script blocking and browser restrictions, increases control over what data the company passes on, and allows it to be adjusted or enriched before sending. It is not, however, a way around consent – if a visitor has declined measurement, it must not be carried out on the server side either. Deployment is more demanding and more costly than classic measurement, requiring its own infrastructure and maintenance. It makes sense for websites with a significant volume of business, where even a few percentage points' improvement in accuracy has value, not for small projects. Before deployment, it therefore pays to calculate what share of business the current measurement fails to capture and whether its value exceeds the cost of running the solution.
See also: Consent mode in analytics, Data quality in analytics, Measuring e-commerce transactions.
