Personalisation of online content is possible without cookies

If you manage a website and want to know as much as possible about your customers, you're probably wondering what to do without cookies in the future. The authorities are pretending that, unlike in the past, active consent to tracking by each user will not only be required, but enforced. This will be the end of cookie-based content personalization, among other things. What are the alternatives?

Behavioural data

It is possible to collect data on users' online behaviour. This so-called "soft" or "soft" data shows, for example, where the user clicked during their visit, what mouse or finger movements they made on the screen, how much time they spent browsing a specific product or just a part of a specific website (e.g. If we let the user actively fill in something on the website, we also add "keystroke dynamics" to the data - i.e. how long it takes the user to fill in a specific field, how many mistakes they make, how many times they use e.g. backspace or delete, etc.

From such metrics, we can compose a behavioral profile of a specific visitor, which we can compare in real time with already collected profiles and calculate the percentage probability that the visitor is a visitor who has already been to the site.

Device detection

Exact information about the device used can be used as additional data to the behavioral profile. This includes the specific make, model, year of release or even commissioning. Then more subtle data such as browser and operating system type, considering the current version, resolution settings, audio, battery status or geolocation data (current approximate location, IP address, etc.).

If we combine all this data, we create a completely accurate "fingerprint" of the device, which will be almost unique in practice. The huge advantage of this approach is, among other things, the fact that the "device data" is available to you instantly (within milliseconds of the user accessing the site) and therefore, you don't have to wait for it to "start behaving" in any way.

Behaviour + device = profile

When we add device data to the behavioural profile of a user described above, the likelihood of two identical profiles occurring within visits to a given site is almost certainly negligible.

Personalizing online content with Zoe.AI

Behavioral metrics (there are more than 300 of them) along with accurate device detection are used by tools such as Zoe.ai. As a result, it can recommend the product the user is most likely to buy in more than 95 % of cases. It recommends individually, i.e. not to a group of users but to a specific individual, without having to identify him by logging in and without using his cookies.

If you want to know how it all works and what exactly it can do, we would be happy to put you in touch with the team that has been developing it for several years. Just drop us a line - perhaps via the contact form.