Server-Side Tracking for GA4: Why It Matters and How to Set It Up

Server-Side Tracking for GA4: Why It Matters and How to Set It Up

ByAdam

Client-side GA4 loses 20–40% of data to ad blockers and browser restrictions. Server-side tracking recovers that data. Here’s how it works and when it’s worth it.

Standard GA4 implementation sends tracking events from the visitor’s browser to Google’s servers. This client-side approach is simple to implement but has an increasingly significant problem: ad blockers, intelligent tracking prevention (ITP) in Safari and Firefox, and privacy-focused browsers block or restrict client-side JavaScript tracking tags. Estimates for how much GA4 data is lost to these restrictions range from 20% to 40%, depending on the audience. For sites with a tech-savvy, privacy-conscious, or adblock-heavy audience, the loss is closer to 40%.

Server-side tracking moves the data collection step from the visitor’s browser to your own server, sending events to Google from a first-party endpoint that isn’t blocked by ad blockers or browser privacy features. The result is more complete data — and with iOS 17’s further restrictions on cross-site tracking, the gap between server-side and client-side accuracy is widening each year.

The Short Version

Client-side GA4 loses 20–40% of events to ad blockers and browser privacy restrictions. Server-side tracking routes events through your own first-party domain before forwarding to Google, recovering most of that gap. It requires a server-side GTM container, cloud hosting (roughly £20–150/month depending on traffic), and a subdomain on your own site for true first-party context. Worth it for sites with ad-blocker-heavy audiences or decisions that depend on accurate conversion data; often not worth the complexity for smaller, lower-risk sites.

DimensionClient-side GA4Server-side GA4
Data completeness60–80% of true traffic85–100%, depending on audience
Setup complexityLow — standard GTM web containerHigher — needs a server container and hosting
Ongoing costNone beyond standard GTM£20–150/month hosting, depending on traffic
Resilience to ad blockersLowHigh, via first-party domain
Best suited forSmall sites, low-stakes dataSites where conversion accuracy drives decisions

How Server-Side Tracking Works

In a client-side GA4 setup: the visitor’s browser loads your page → the browser executes the Google Tag Manager (GTM) container → GTM fires the GA4 tag → the GA4 tag sends a hit to Google’s servers. Ad blockers intercept the GTM request or the GA4 hit and block it.

In a server-side GTM setup: the visitor’s browser loads your page → the browser sends events to your server-side GTM container (hosted on your domain) → your server-side container forwards the events to GA4. Because the event goes from the visitor’s browser to your own domain (first-party), ad blockers that allow first-party requests don’t block it. The final forwarding step (from your server to Google) also isn’t blocked because it happens server-to-server, outside the browser.

What You Need to Implement Server-Side Tracking

  • A server-side GTM container: Google Tag Manager offers a server container type (separate from the web container). You need both: the web container sends events to the server container; the server container processes and forwards them to GA4.
  • A cloud server to host the server-side container: Google Cloud Run, AWS, or DigitalOcean work. Google Cloud is the easiest setup via the GTM server container wizard. The server runs 24/7 and adds a hosting cost (typically £30–120/month depending on traffic volume).
  • A subdomain on your own domain: The server-side container should be served from a subdomain of your site (e.g., `data.yourdomain.com`) to ensure first-party context. Using Google’s default tagging.example.com endpoint doesn’t provide the first-party benefit.
  • Updated web container configuration: Your existing GTM web container needs to be configured to send events to the server container URL rather than directly to GA4. The GA4 Configuration tag in your web container points to the server container endpoint.

When Server-Side Tracking Is Worth the Cost

The added complexity and hosting cost are worth it when:

  • Your audience has high ad blocker adoption (tech, developer, privacy-focused communities)
  • Your decisions depend on accurate conversion data (paid advertising, SEO investment decisions based on organic conversion rates)
  • You need accurate revenue attribution from ecommerce purchase events
  • You’re running remarketing campaigns that depend on complete audience data

For smaller sites with lower-risk audiences and less reliance on precise conversion data, client-side GA4 with consent mode v2 is a reasonable approach. The data gap is real but may not justify the implementation complexity and ongoing cost until the site reaches a scale where data accuracy meaningfully changes decisions.

20–40%
Of client-side GA4 events typically lost to blocking
25–40%
More sessions recovered for ad-blocker-heavy audiences
8–20h
Typical developer time for a standard implementation

For the broader GA4 measurement setup, see GA4 for SEO. For how conversion event accuracy affects SEO reporting, see conversion events that tie to revenue.

A Worked Example

A B2B developer tools company suspected its GA4 conversion data was understating performance — internal CRM numbers showed roughly 30% more sign-ups than GA4 was reporting as conversions, and the audience (developers, many running ad blockers and privacy extensions by default) was an obvious candidate for high data loss.

The team set up server-side GTM on Cloud Run with a dedicated first-party subdomain, running it in parallel with the existing client-side implementation for four weeks to quantify the actual gap before fully switching over. The parallel run confirmed a 34% under-count in client-side GA4 relative to the server-side numbers, closely matching the CRM discrepancy that had originally raised the question.

After switching primary reporting to the server-side data, the team’s organic conversion rate — previously calculated using the under-counted client-side numbers — increased by roughly a third overnight, with no actual change in site performance. This materially changed the calculated ROI of the SEO programme in the next budget review, simply by correcting a measurement gap rather than improving anything about the site itself.

Assess Your Tracking Accuracy

Frequently Asked Questions

No — GDPR compliance is about consent and data processing, not the technical method of data collection. Server-side tracking still requires valid user consent for analytics and advertising purposes under GDPR. What changes with server-side is that you have more control over what data is sent to Google (you can strip PII, aggregate data, or apply transformations server-side before forwarding to GA4) — which can actually make compliance easier to manage than client-side. A consent management platform (CMP) that implements Google’s Consent Mode v2 is still required; server-side tracking doesn’t bypass consent requirements, it just provides more reliable data collection for consented users and fewer accidental PII leaks.

Recovery rates vary by audience. In documented implementations across digital marketing agency sites and developer-focused tools (audiences with high ad blocker rates), server-side tracking recovers 25–40% more sessions than client-side GA4 alone. For general e-commerce and consumer service sites, recovery is typically 15–25%. After implementing server-side, run both implementations in parallel for 4 weeks to measure the actual gap for your specific audience — the aggregate figures are a useful benchmark but your audience may differ significantly from the average.

It’s a complement, not a replacement — in most implementations, both client-side and server-side tags run together. The client-side web container still loads (for events where client-side data is needed, like scroll depth and click interactions), but it points to the server-side container rather than directly to GA4. The server-side container then forwards to GA4 and any other marketing platforms. Some advanced implementations run the web container in minimal mode (just capturing events and forwarding them) with all processing done server-side — this is more privacy-preserving but also more complex. For most businesses, the hybrid approach (standard web container + server-side forwarding to GA4) is the right balance of completeness and complexity.

The primary cost is cloud hosting for the server-side GTM container. On Google Cloud Run, a site with up to 100,000 monthly visits can typically run server-side GTM for £20–50/month. Higher-traffic sites scale linearly: 500,000 visits/month is approximately £60–150/month. These are running costs; the one-time implementation cost (developer time for setup and configuration) typically ranges from 8–20 hours for a standard implementation, or £800–3,000 if done by an agency. The implementation cost is usually recoverable within the first year if the improved data quality changes even one significant campaign or investment decision.

Compare your GA4 conversion counts against a source of truth you trust — CRM lead records, ecommerce order data, or email confirmations sent on form submission. If GA4 is undercounting by more than roughly 15–20% against that source, the gap is large enough that server-side tracking is very likely worth investigating further. If the gap is small (under 10%), the cost and complexity of server-side tracking probably isn’t justified yet, and the money is better spent elsewhere until the site’s traffic or stakes grow.

A developer or technical marketer comfortable with GTM and basic cloud infrastructure should own it, since the container needs occasional updates (new tags, GTM platform changes, certificate renewals on the hosting side). It’s not a “set and forget” system — budget a few hours per quarter for maintenance and monitoring, and make sure whoever owns it has access to both the GTM account and the cloud hosting console, not just one or the other.

Better Data Changes Better Decisions

Analytics decisions made on data that’s 20–40% incomplete aren’t just imprecise — they’re systematically biased. If your client-side GA4 under-reports conversions by 30%, your organic search conversion rate looks worse than it is, your SEO investment case looks weaker than it should, and your attribution to paid channels is inflated (because ad platforms use their own conversion tracking, which doesn’t have the same blocking problem). Server-side tracking is about getting the accurate picture on which real business decisions rest.

If you’d like help evaluating and implementing server-side tracking for your site, get in touch.

Similar Posts