Cross Devices Tracking

User tracking is almost as old as the Internet itself and is used on many websites today – be it “only” Google Analytics for tracking user behavior and traffic. Especially in the e-commerce sector, user tracking is a popular and proven method for analyzing the purchase and cancellation behavior of customers and accompanying them with appropriate (often automated) measures. If a user cancels the purchase at a certain step in the shopping cart, a voucher code for exactly the item in the shopping cart can be generated via dedicated tracking and displayed on other (external) websites, thus moving the user to still make the purchase at a later point in time.

Furthermore, such tracking is also used for cross-selling and up-selling. For example, if a customer buys a digital SLR camera online, the shop can offer him suitable accessories such as a camera bag, memory cards or suitable lenses a little later.

Amazon Brand Guide

Cross Device Tracking instead of classic cookies

However, the procedure described above is increasingly reaching its limits because it is based on cookies. A small tracking text file (called a cookie) is stored on the user’s computer so that his customer journey can be traced accordingly. Most affiliate programs (e.g. Amazon PartnerNet) are also based on this procedure in order to reward brokered purchases with commissions.

The problem: Cookies are permanently stored on a device and more and more users are not only using a (fixed) device in the network, but also use different end devices for surfing. If a user now buys a product with one device (e.g. laptop) and surfs on later with a second one (e.g. smartphone), the cookie tracking practically no longer works correctly. This is exactly where cross device tracking comes in, which can function in three different ways.

Cross Device Tracking with User ID

The “easiest” method of tracking a user without cookies is to use a user ID. An existing login is often used as a basis to clearly identify a user. Social networks such as Facebook, Twitter or Google play an important role here, because users are often logged in on several different devices with the same account and can thus be clearly “marked”.

Cross Device Tracking with Device ID

Another method of cross device tracking is the use / generating of a device ID. Since this method is much more inaccurate than the identification using cross-device accounts, much more data must be collected and used for comparison when setting up a device ID. For example, installed apps, used browsers, saved bookmarks or used IP addresses (e.g. same WLAN networks) are used to create a user profile.

Cross device tracking with probabilistic algorithm

Another possibility of cross device tracking is the use of probabilistic algorithms. For this purpose, a kind of user DNA based on probability calculation is practically created. In order to generate such a “DNA”, many different data from surfing behavior and browser history are used. Every user moves through the Internet with a certain pattern and can be identified with this method. The accuracy varies between 70 and almost 100 percent (source:

Privacy and User Tracking

As useful as it can be for advertisers and online sellers to identify visitors and ideally turn them into paying customers, user tracking is problematic in terms of applicable data protection regulations. According to this, the collected data must not be directly and unambiguously assigned to a human being, but must always be stored and further processed anonymously. This in turn complicates user tracking and makes it almost impossible. Especially with the Basic Data Protection Regulation 2018, the rules for tracking will be tightened again. Before using any tracking method, it is advisable to seek professional (technical and legal) advice in order not to violate applicable law.