In a constantly evolving tracking landscape with Apple’s regular ITP updates, the impending end of third-party cookies, and the instability of data transfer relations between the USA and the European Union, many advertisers have implemented a server-side tracking solution. Have you just implemented it or are you planning to do so in the near future? You’re probably wondering how to measure the results and justify the implementation of this technology.

In this article, you will find the processes to establish in order to visualize the increase in your results following the migration.

I- Reminder of the Interest of Server-Side Tracking

First, let’s recall what server-side tracking with Google Tag Manager is. This technology, which communicates with your classic GTM container (called client-side), allows you to implement specific APIs and collect the best data. If you want to learn more, don’t hesitate to check out this article: What Is GTM Server-Side, and Why Should You Use It?

Implementing such a stack will enable numerous possibilities, including:

• Bypassing ad blockers
• Circumventing Apple’s ITP restrictions on iOS and Safari
• Implementing Meta CAPI, Pinterest CAPI, and TikTok Events API
• Proxyfying your analytics tool to control and transform the data sent
• Improving page load times by moving many client-side scripts to the server

After implementing these actions, it’s natural to want to observe the increase quantitatively. That’s why you’ll find the methods below to compare results between client-side and client-side + server-side.

II- The Impact of Ad Blockers

It is estimated that about 30% of French users use ad blockers (source). These tools offer ad-free browsing but can also block essential trackers for advertisers. By adopting server-side tracking, the impact of ad blockers on data collection can be eliminated. Addingwell’s autoOps solution, which handles server hosting, provides a CDN to proxyfy without a client-side GTM container, allowing you to bypass these blockers.

To measure the impact of this bypass, there are two main methods:

1. Use Simo Ahava’s method with a script: This clever method from Simo Ahava, one of the global tracking experts, allows you to detect different levels of blocking (Google Analytics, GTM, and ads) on the client-side and transmit this data to the server. From there, you can transfer these events and dimensions to Google Analytics 4 (or any other analytics tool like Piano or Matomo that also have server-side tags).
2. Set up a second GA4 property: This method involves creating a second Google Analytics 4 (or another web analytics tool) property as a control. For testing purposes, I recommend this method, which also allows you to measure the impact of ITP (see below). However, note that this may not be ideal in the long term as it negatively affects page load times by adding an additional script.

To implement this, you must do the following in addition to your classic client + server GA4 setup:

a) Configure a new GA4 property.

b) Insert the GA4 script directly on the website.

I emphasize that you should not add this property to your client-side GTM container, as it would bypass ad blockers and render the test useless. You can then observe the difference in data volume between your two properties and measure the impact of ad blockers.

III- The Impact of ITP

Apple’s Intelligent Tracking Prevention (ITP) on iOS and Safari limits the lifespan of first-party cookies (in addition to preventing third-party cookie placement). This limitation on lifespan (either one day or seven days depending on the acquisition source) significantly affects the attribution of your marketing campaigns for both analytical and marketing purposes in advertising platforms.

If ITP doesn’t limit your data collection volume, it will impact its quality. To bypass this limitation, you can use the tools provided by Addingwell (source). You can then use the Google Analytics 4 test property to compare:

• The number of new users: This figure should be lower on your server-side property.
• User acquisition sources: The direct source should logically decrease on your server-side property.

IV- The Impact of Conversion APIs

Conversion APIs, such as Meta CAPI, Pinterest CAPI, and TikTok Events API, offer significant performance advantages. From my experience, I observe a 20% increase in event collection. This number can vary depending on your vertical.

For Meta (and this works for other platforms as well), you need to go to the event collection report in the event manager > data sources.

You can then see the difference in data received by the pixel and the conversion API.

Another interesting indicator provided by Meta is the number of additional conversions recorded. Beyond the number of events received by the platform, this figure indicates the additional conversions attributed to the Meta Conversion API installation. This figure is updated once every seven days, and if you want to learn more about this indicator, I invite you to consult the official documentation (source).

Here, we observe an increment of 44.5% in conversions from the Conversion API compared to the data collected by the pixel.

Please be cautious about the advanced event matching rating. There are four rating levels:

• Poor
• Okay
• Good
• Excellent

However, these ratings do not reflect the quality of the events received but only the user data you can send to the platforms (source). Additionally, this indicator is calculated based on an average of other advertisers and does not indicate the actual relevance of the data you may have sent and matched with Meta’s database.

The Meta Conversion API is undoubtedly one of the use cases that most strongly encourage advertisers to switch to server-side tracking, and these numbers alone can justify the investment if ad spend volumes are significant. Having more events not only improves ad algorithm optimization but also provides more complete re-targeting audiences, potentially leading to better performance.

V – Impact on Web Performance

Another significant advantage of server-side tracking is the reduction in website load times. Indeed, removing client-side scripts for server-side transfers automatically leads to a gain in page loading speed. This allows you to transfer tags from Pinterest, Meta, Google Ads, Outbrain, Taboola, and other affiliate platforms to your GTM Server-Side container.

To measure the impact, you have no choice but to establish baseline figures before implementation and compare them after the implementation.

You can use specific tools for measuring page load times, but be sure to measure the entire website comprehensively and avoid focusing solely on a specific page to avoid biasing your analysis.

You can also set up tracking of page load times via Google Tag Manager one month before your migration. This will provide you with fairly comprehensive statistical data. You can use this template (developed by Simo Ahava once again): Track Core Web Vitals in GA4 with Google Tag Manager, which aims to collect various Core Web Vitals indicators (LCP, FID, CLS) in your analytics tool.

Following the implementation of server-side tracking via GTM, you can then measure the differences in loading times between your initial implementation and the new one.

The migration to server-side tracking offers numerous advantages in terms of tracking, overcoming limitations, and improving web performance. To justify these installations, it is essential to implement appropriate measures before and after implementation to visualize these benefits quantitatively. This way, you can demonstrate the value of this solution in the long term.

Conclusion

The migration to server-side tracking offers numerous advantages in terms of tracking, overcoming limitations, and improving web performance. To justify these installations, it is essential to implement appropriate measures before and after implementation to visualize these benefits quantitatively. This way, you can demonstrate the value of this solution in the long term.