Server-side tracking improves data quality but does not replace consent. Anyone processing user data for advertising purposes in the EU generally needs valid consent, regardless of whether data is transmitted from the browser or via a server. A good measurement plan therefore combines clean consent management, server-side transmission for users who have consented, the platforms' conversion APIs and an overall view based on shop data that does not rely on individual user tracking.
Why tracking data has gaps
Data in ad platforms and analytics tools is incomplete for several reasons:
- users decline consent or ignore the banner
- browsers limit cookie lifetimes and block certain scripts
- ad blockers suppress tracking scripts
- users switch between devices
The result: platforms see only part of the conversions, and algorithms learn from a narrower data set. At the same time, platforms claim conversions that would have happened without advertising. Both effects make any single data source unreliable.
The legal basis in brief
In Germany, Section 25 of the TDDDG governs access to end devices, such as setting cookies or reading identifiers. This generally requires consent unless access is strictly necessary for a service the user has requested. The GDPR applies to the subsequent processing of personal data. For marketing and analytics purposes, consent is in most cases the viable legal basis. Comparable rules apply across the EU.
This article does not replace a legal review. Have your setup assessed by a qualified adviser.
What server-side tracking does
With server-side tracking, the browser does not send events directly to many third parties but to your own server endpoint. From there, data is forwarded to platforms in a controlled way.
Advantages:
- Control: you decide which data goes to whom and can remove fields before forwarding.
- Data quality: events arrive more reliably because fewer scripts run in the browser.
- Page speed: fewer third-party scripts lighten the page.
- Enrichment: information such as margin or new-customer status can be added server-side, enabling steering by contribution margin.
What it does not do: it does not make missing consent unnecessary. Using server-side tracking to bypass consent creates a legal risk, not one data problem fewer.
The platforms' conversion APIs
Meta, Google, TikTok and others offer interfaces through which conversions are sent directly from the server or the shop. What matters:
- Deduplication: if browser and server send the same event, the platform must recognise it as one. That requires a shared event ID.
- Event quality: the more permissible, hashed customer data (where consent exists) is sent, the better the platform can match.
- Consent status: the consent status must be transmitted and respected.
Consent management that works
A consent banner is not purely a legal topic; it directly affects your data. Compliant banners offer an equivalent option to decline. Within that framework there is room for manoeuvre: clear language, clear purposes, a calm design that fits the brand, and a technically clean implementation in which no marketing scripts load before consent.
Check regularly whether newly added apps or scripts bypass the banner. It happens more often than you would think.
Modelling and the overall view
Because individual user data will stay patchy, methods that do without it are gaining importance:
- MER and total revenue: how is shop revenue developing relative to marketing spend?
- Incrementality tests: pause advertising in a region or for an audience and measure the difference.
- Marketing mix models: statistical models based on time series, useful from a certain size and data history.
- Platforms' modelled conversions: helpful, but to be understood as estimates.
Checking data quality
- Compare shop orders with the conversions in each platform for the same period
- Check deduplication: are purchases counted twice?
- Check events on test orders, including values and currency
- Monitor consent rates, especially after changes to the banner
- Document and explain discrepancies
A realistic measurement plan
| Level | Tool | Purpose |
|---|---|---|
| Company | shop and accounting data, MER | the truth about revenue and efficiency |
| Channel | platform data with conversion API | day-to-day steering |
| Validation | incrementality tests | checking what is really additional |
| Analysis | privacy-compliant web analytics | understanding on-site behaviour |
Checklist
- consent banner reviewed technically and in content
- no marketing scripts before consent
- server endpoint with clear data forwarding
- conversion APIs with deduplication
- regular reconciliation with shop data
- MER as the overarching metric
- legal review of the overall setup
More on budget steering on this basis in Planning a Marketing Budget and Google Shopping and Performance Max. If you want to run this with your own numbers, I can help in performance marketing consulting.
How does your ad account compare?
Describe your channels and budget in a few lines. In the initial call we work out what you should steer by.
Book an initial call E-Commerce & MarketingFAQ
Is server-side tracking allowed without consent?
For marketing and analytics purposes, generally not in the EU. Whether consent is required depends on the purpose and on access to the device, not on the transmission route. Have your specific case reviewed legally.
Why do Meta and Google show more conversions than the shop?
Because they attribute conversions by their own rules, partly based on views rather than clicks, and purchases overlap. The sum of platform figures is therefore often higher than actual revenue.
Is a marketing mix model worthwhile for smaller retailers?
Usually only from a certain size, variety of spend and data history. For smaller retailers, MER and simple incrementality tests are often the more pragmatic route.
What should the consent rate be?
It varies greatly by audience and design. More important than a target is a compliant design and tracking changes over time.