Nowadays, media content is consumed via desktops, laptops, tablets, smartphones or even smart TVs. A single user often uses multiple devices and switches seamlessly between them. This device multitasking makes it difficult for advertisers to apply their advertising campaigns efficiently.
To be able to trace the entire customer journey of a user, several devices must be assigned to the individual user.
HAVE COOKIES BECOME OBSOLETE?
In fact, studies show that more than half of all customer journeys take place on more than one device. However, the assignment of users to different devices does not work via cookies, as these do not identify the user itself, but solely devices and browsers. In addition, cookies are regularly deleted and therefore cannot provide relevant results over a long period of time. In order to understand the Customer Journey from the first touchpoint to the conversion and be able to include all touchpoints, cross-device tracking must be used.
Statistics show that insights from cross device attribution can reduce the CPA by 30% to 50% and even increase the ROI by 50%. Using a cross-device view, consumers also convert 1.4 times more often than with a single-device view, as they are granted a coherent, personalized arrangement across all channels.
HOW DOES CROSS DEVICE TRACKING WORK?
For cross-device tracking, service providers use a deterministic or a probabilistic model. Choosing the right model is crucial for the accuracy of the results.
THE PROBABILISTIC MODEL:
The data in the probabilistic model is based on anonymized user profiles, which are generated by statistical or machine learning methods and determine probable connections. The user behavior is analyzed on different devices with the help of different data points from which a pattern is then formed. This creates a kind of user DNA, which can later on be synchronized on different devices. This matching technique relies heavily on probability principles and is therefore very imprecise.
THE DETERMINISTIC MODEL:
In deterministic models, first-party data (usually login data) is used to create a link between a user and his devices. For example, if a user logs on to an advertiser's website from different devices, it is clear that the devices in question belong to the same user. The deterministic matching does not work with algorithms based on probability assumptions to link a user to his devices. Instead, this form of matching requires a clear connection between a user and a device. The advantage of this form of tracking is the high accuracy. It is a lot more precise than probabilistic tracking, since a device is only connected to a user if this has been distinctly matched beforehand.
WHY IS THE DETERMINISTIC MODEL SUPERIOR?
The deterministic model is particularly accurate and clearly more precise than the probabilistic model. A mapping between user and device is only achieved if there is a clear connection between the two. The probabilistic model, on the other hand, is based on algorithms and probability assumptions, so that the hit rate is much lower. Still, probabilistic matching allows a greater range than deterministic matching, since it does not depend on users to perform certain actions (such as logging on to a device). However, this method is less precise than deterministic matching: hit rates vary greatly and in some cases are as low as 60%.
CROSS DEVICE POOLS - BENEFITS AND PRIVACY COMPLIANCE
Customer journeys can be completed by incorporating findings already gained from existing, anonymous data from comparable advertisers. For this purpose, companies share their cross-device data with each other in order to benefit from a large data pool.
The data is made available in an encrypted form so that it cannot be decrypted. In this way, the user data remains protected and can be used for cross-device analysis. This ensures that cross-device tracking complies with data privacy and the basic data protection regulation (DSGVO).
Cross-device attribution provides an accurate pattern of user behavior and helps to design a customer journey with all its user touchpoints. This is also shown by an internal study by Exactag, in which 62% of additional mobile touchpoints could be identified and assigned to existing journeys. Through a common cross device pool of anonymized user IDs, exactag customers not only receive clear hit rates, but also maximum reach.