What will targeting look like in a world without third-party cookies?
Google Privacy Sandbox
One of the leading alternatives comes from Google. With its ‘Privacy Sandbox’, Google seeks to help develop as many different approaches as possible. The results should then lead to open standards. For this reason, the World Wide Web Consortium (W3C) is involved in the project.
The list of ideas is long and contains vastly different concepts for various fields of application. The biggest challenge faced by the Privacy Sandbox: It takes a lot of time to develop a universally recognised standard. There are also several drawbacks, like with the FLoC model (Federated Learning of Cohorts). This model was the first choice for a while, but was ultimately rejected by too many data protection bodies and browser developments.
The Topics API provides one example of this. The idea behind this API is to enable ads to be aligned to user interests while simultaneously protecting their privacy. The concept: A website assigns its content to generic lists of topics like ‘Football’ or ‘Entertainment’. The user’s browser then determines which 5 topics the user is most interested in at present. Users can view and modify this list, or opt out of participating in the Topics API.
The FLEDGE API (First Locally-Executed Decision over Groups Experiment)
takes another approach. In this case, the idea is to move the bidding process for ad space from numerous online servers to the browser itself. Ad customers should still be able to reach certain user groups. Retargeting and frequency capping would likewise be possible. Ultimately, ads should only be transmitted by trustworthy servers.
As things currently stand, it remains to be seen which ideas will succeed and
what they will achieve in practice.
One thing that is somewhat easier to predict is a return to first-party data.
Everything website operators stand to learn about their users themselves won’t be affected by the upcoming changes.
This includes information from web analytics, data from orders and queries along with additional information voluntarily provided by customers and users.
All this data needs to be collected and processed in compliance with data protection regulations. In addition, users need to be clearly told which data is collected, how it’s protected against unauthorised access and which purposes it serves.
Another challenge lies in encouraging users to provide more information. Typical ways to achieve this include competitions, gated content, newsletters and events.
Customer Data Platforms and Data Clean Rooms
Reviving classic targeting methods
Proven methods are making a comeback. Contextual targeting is back on the scene, whereby ads are aligned with the content of a page. Google Ads has successfully followed this approach for a number of years.
Affinities offer another approach. Instead of revolving around matching keywords, they focus on general interests and characteristics. For example: Someone researching racing bikes is likely to also be interested in healthy nutrition. Other data such as age group can also be derived subject to certain limits depending on the website.
Conventional advertising methods also cover universal IDs such as Unified ID 2.0. With universal IDs, the sole focus lies in motivating users to register on a website. In this way, they voluntarily identify themselves. In return, they can configure their ad settings, which automatically apply to any site using the ID provider. However, some analysts claim only around 20% of users can be reached in this way.
To be well-prepared for a future without third-party cookies, companies need to be acutely aware of the needs of their target group, what motivates these customers and which websites, platforms and channels are best for reaching them. As mentioned above, first-party data plays a key role in this.
If tools like retargeting disappear, companies may benefit from acknowledging the renewed importance to asking questions about their own websites: What is causing visitors to leave our website? How can this be avoided from the outset?
All this will require more work by everyone involved over the coming years. That being said, their efforts will likely pay off several times over when companies are able to collect their own data, gain a better understanding of their customers and optimise their services in line with customer expectations.
About Kai Vorhölter: