Marketing analytics when cookies can’t be trusted: why server-side tracking matters

Updated on October 18, 2026: figures reviewed and updated.

For years, marketers prepared for the day Chrome would drop third-party cookies. That day never came. On April 22, 2025, Google announced it would keep the current approach: users decide in Chrome’s settings whether to block or allow third-party cookies (Google, April 2025). On October 17, 2025, Google went further and retired most Privacy Sandbox technologies, including Topics, Protected Audience and Attribution Reporting (Google, October 2025). Cookie-based analytics data still has gaps, for other reasons. Safari blocks third-party cookies outright and caps how long cookies set by JavaScript can live, Firefox blocks cookies from known trackers, and ad blockers stop measurement tags from firing at all. This article walks through the key milestones, three common analysis errors, and how server-side tracking reduces the share of data you lose.

Timeline of cookie restrictions

These are the main milestones in cookie restrictions across the most widely used browsers and operating systems.

1. In 2017, Apple launched Intelligent Tracking Prevention (ITP, an anti-tracking feature) in Safari, switched on by default for every user, to restrict cookies set by cross-site trackers. Starting with ITP 2.1 (February 2019), cookies created by JavaScript were capped at a maximum lifetime of 7 days.
2. In June 2019, Firefox turned on Enhanced Tracking Protection by default, blocking cookies from known third-party trackers (first for new users, then for existing ones).
3. On March 24, 2020, Apple (WebKit) announced full default blocking of third-party cookies in iOS and iPadOS 13.4 and in Safari 13.1 on macOS. Since 2019, first-party cookies created by JavaScript (such as the Google Analytics cookie) had already been capped at 7 days, and in some cases at just 24 hours.
4. On June 22, 2020, Apple announced that apps on iOS 14 would have to ask users for permission before tracking them or accessing the device’s advertising identifier (IDFA). In April 2021, after a public standoff with Facebook, the company released App Tracking Transparency (ATT) to all users in the iOS 14.5 update. Earlier restrictions only touched web analytics. Since ATT, mobile analytics has become much harder too.

What about Chrome?

In January 2020, Google said it would phase out third-party cookies in Chrome within two years, then pushed the deadline back several times. In April 2025, Google decided to keep third-party cookies (with users choosing in their settings), and in October 2025 it shut down most Privacy Sandbox technologies. Even so, Safari, Firefox, ad blockers and legal consent requirements still leave cookie-based data with large gaps.

How these issues affect marketing analytics

The chart below shows US browser market share according to Statcounter at the time the original article was written:

In practice, this means you may have been making decisions on inaccurate data, because the browsers listed above interfere with how cookies work. For readers in Vietnam, Safari accounts for about 20.6% of web browsing (StatCounter, September 2026), so the share of data affected by ITP is still large even though Chrome kept third-party cookies. Picture how much that complicates marketing analytics.

Here are three examples of analysis problems marketers run into because of third-party cookie restrictions.

Case one

The problem: referral traffic disappears and direct traffic goes up.

What actually happened: Imagine a user whose main browser is Safari. He visits three times from ad channels, then comes back to the site from a bookmark 7 days later and places an order. Google Analytics identifies him as a new user and records his source as direct/none.

The takeaway: You cannot tell which ad channel the customer came from, and the share of direct visits to the site goes up.

Case two

The problem: attribution models keep getting less accurate.

Multi-channel attribution matters a great deal in e-commerce and SaaS. Before a user completes a target action, such as a purchase or a sign-up, they usually interact with several different ad channels along the way. Marketing’s job is to split the budget sensibly across those channels and get the highest possible ROAS (a metric for how profitable ad spend is).

The takeaway: in reality, attribution built on incomplete data is the result of losing part of the chain of ad channels. That makes it very hard to follow the customer’s full path.

Case three

The problem: reports based on cohort analysis stop adding up.

What actually happened: In the Retention report, your cohorts look weaker than they are. Because of user identification problems, returning users get counted as new ones.

The takeaway: reports built on third-party data show a smaller real user flow than you actually have.

Why these cases matter

Reliable data collection is the base of any data-driven marketing strategy. Shorter cookie lifetimes hit data quality and decision-making directly. The fix is to move away from third-party cookies toward server-side tracking built on first-party data.

Client-side vs. server-side tracking: what is the difference?

There are two ways to measure how users interact with a website. The difference lies in how the data is collected.

Client-side tracking

With client-side tracking, data goes straight from the user’s browser (the visitor’s machine) to an outside server such as Google Analytics or Facebook Pixel. It travels through small snippets of JavaScript placed on the site, usually called pixels or tags.

Server-side tracking

Server-side tracking lets you send raw data from the user’s browser to your own web server and receive it there before it goes anywhere else. Adding that extra layer (a server) between your website and data collection platforms such as Google Analytics or Facebook gives you more security and more control over the data.

Client-side tracking was the preferred way to collect data for a long time, but that is starting to change. More and more marketers favor server-side tracking, because ad blockers and intelligent tracking protection (ITP) in Safari and Firefox have no effect on it.

Main pros and cons of server-side and client-side tracking

Server-side

• Higher data quality, because ad blockers and intelligent tracking protection (ITP) do not affect it.
• Control over and ownership of your data: you decide what to track and where to send it.
• Faster websites, since page load time goes down.
• Richer data: you can add relevant extra information, such as data from your CRM.
• The data belongs to you.

Client-side

• Ad blockers and ITP (Intelligent Tracking Protection) block data collection and lower data quality.
• It slows the website down.
• You do not own the data. If you use Google Analytics, Google Analytics stores and owns it.
• Data security concerns put limits on what you can collect.

Getting marketing analytics ready as cookies become less reliable

Renta gives data-driven teams access to customer data within minutes. It delivers data automatically and in real time from closed sources to destinations such as Google BigQuery or the Facebook Conversions API.

Server-side tracking with Renta

• Collects product and marketing analytics data into a data warehouse, in a data format that follows the Google Analytics 360 standard.
• Native server-side integrations with third-party marketing and sales tools.
• Compliant with privacy rights, GDPR and Intelligent Tracking Prevention.

How does server-side tracking work?

You create messages and send them to the Renta API in JSON format. The API then converts the data into whatever format third-party tools and data warehouses need.

Marketing ETL

Marketing ETL lets analysts pull data into their own warehouse in just a few minutes. Renta has more than 30 ready-made integrations.

• Flexible configuration of the data structure.
• Automatic updates and timely data delivery.
• Backfill of historical data updates.

How to switch to server-side tracking

What server-side tracking looks like in practice

1. You set up the events to be recorded on the website, for example one page view or one transaction. The setup is simple and no different from installing Facebook Pixel.
2. All events are then collected and sent to your dedicated server infrastructure. On the server, the data is standardized and converted into various formats for tools or data warehouses (together called destinations).
3. Through a simple interface, you can configure other integrations that pass the collected data along in real time, for example to Google BigQuery, the Facebook Conversions API or Salesforce.

Every available integration with third-party applications is implemented only as a server-side integration.

Adapted by Adtek from original content by Renta, updated on October 18, 2026.