E-commerce tracking: stop advertising blind

Most webshops spend more money on ads than on anything else, and yet the tracking behind that money is often leaky. The result is that you scale what looks best in a dashboard, rather than what earns the most in reality. Precise e-commerce tracking and well-considered attribution are what separate guessing from knowing.

Why tracking has become hard

There was a time when a tracking script in the browser caught almost everything. That time is over. Cookie consent, browsers that block third-party cookies, ad blockers and iOS updates have left holes in classic tracking. A large share of your conversions go unrecorded, or get recorded in the wrong place.

The consequence is serious. When data is incomplete, both you and the ad algorithms make decisions on the wrong basis. That is why control over your marketing spend starts with getting the measurement itself right.

Server-side tracking

The solution to much of this is to move tracking from the browser to the server. With server-side tracking, data is sent from your own server to the platforms, instead of the browser doing it directly. That makes measurement more robust against blocking and cookie restrictions, and it gives you control over which data goes where.

A typical setup uses a server-side container, often in Google Tag Manager, combined with the platforms’ own APIs, such as Meta’s Conversions API and Google’s Enhanced Conversions. It is more technically demanding than a script in the browser, but the payoff is that your conversions get counted.

Attribution models

Tracking tells you that a conversion happened. Attribution tells you who deserves the credit. That is a harder question than it sounds, because most customers see several channels before they buy. Perhaps a Facebook post, then a Google search, and finally an email.

Last-click attribution gives all the credit to the final click. It is simple, but it systematically undervalues the channels that create demand early in the journey. Data-driven attribution, which GA4 offers, instead distributes credit across the touchpoints based on real patterns. No model is perfect. The point is to choose deliberately and understand what your model highlights and hides, instead of taking the default setting for granted.

GA4 as the foundation

Google Analytics 4 is built for the new reality. It is event-based rather than session-based, it supports data-driven attribution, and it can be fed server-side. Set up correctly, GA4 is the backbone of your e-commerce tracking.

That does require the e-commerce tracking to be implemented properly, with the right events for product views, add to cart and purchase, and with correct values on each. A half-implemented GA4 gives you numbers you cannot trust, and numbers you cannot trust are worse than no numbers.

Consent and lawful tracking

Tracking has to be lawful before it can be effective. A cookie consent setup that meets the rules is more than a legal duty, it is part of the setup. With Google’s Consent Mode you can respect the user’s choice and still model the conversions that are lost when someone says no to cookies. The result is a data foundation that is both accurate and defensible. Building tracking that ignores consent is short-sighted. It exposes you and undermines the trust a webshop lives on.

Build one shared data foundation

As traffic grows and channels multiply, a new problem appears: every platform tells its own version of the truth. Meta takes credit for sales, Google does the same, and added together you are selling twice as much as you really are. The solution is to gather your data in one place, often in a data warehouse, where you can hold the platforms’ numbers up against your actual order data from the webshop. Then you have one number to steer by, instead of five pointing in different directions. That is the foundation for all serious cross-channel analysis.

Test your tracking before you trust it

A tracking setup is never something you build once and forget. It breaks, often without anyone noticing. A developer changes something on the checkout, a plugin gets updated, and suddenly half your purchases are no longer counted. So test the tracking regularly. Buy something on your own shop and follow whether the event makes it all the way to GA4 and the ad platforms with the right value. Hold the platforms’ reported conversions up against your actual order data in the webshop. If there is a large gap, something is wrong, and it should be found before you make more decisions on the numbers.

From data to controlling spend

Tracking and attribution are never the goal in themselves. The goal is to steer the money by what genuinely works. When the data is precise, you can move budget from channels that merely harvest existing demand to the ones that create new demand, and you can see what a customer really costs across the whole journey.

The strongest step is to feed the platforms profit rather than revenue. When the algorithm optimizes toward what you genuinely earn, instead of pure revenue, it changes what gets scaled. That is the whole idea behind POAS, and it presupposes that the tracking is in order.

Control starts with clean data

You cannot control what you do not measure precisely. Effective e-commerce tracking and well-considered attribution are the unglamorous foundation that makes all other optimization possible. Because I run webshops myself as CEO, I always build that foundation first. Read more about how I think across channels under omnichannel, or book a meeting and we will look at your tracking together.