Traditional tracking in the user’s browser (Client-Side) is dying. With the advent of adblockers, Safari’s ITP protection (which deletes cookies in 7 days), and iOS restrictions, companies are losing between 30% and 40% of their measurement data.

The technical answer to this problem is Server-Side Tracking.

1. The fundamental difference

Moving your tracking to the server means taking control of the data flow before sending it to third parties:

  • Client-Side (Traditional): The user’s browser downloads scripts from Google, Meta, and TikTok and sends data directly to their servers. It’s heavy, slow, and easy to block.
  • Server-Side: Your website sends a single signal to your own cloud server (a Google Tag Manager Server-Side container). Your server cleans the data and distributes it through the back door to Facebook, Google, and TikTok.

2. Interact: Client vs Server-Side Flow Simulator

Switch between the tabs to see how requests behave in the user’s browser and how privacy blocks are bypassed.

Browser sending data (Client-Side)

The browser makes multiple calls directly to third-party domains. Adblockers detect and block half of the requests.

User Browser

3. Conversions API (CAPI) and Deduplication

The Meta Conversions API (CAPI) is the perfect complement to the browser pixel in server-side tracking. The server connects directly with Meta’s servers via API.

However, to prevent Meta from counting sales twice (duplicated), it is mandatory to set a unique event_id on both the pixel and the server. Meta will read the ID and, if they match, will automatically deduplicate them.