This project was conducted in 2023 as part of our framework agreement with the Directorate for Government Transformation (DITP). We assisted the DITP behavioural sciences department following a solicitation from the French administrative regulatory body whose mission is to ensure that data privacy laws : the Commission Nationale de L’informatique et des Libertés (CNIL).
In 2018, the European Union introduced the General Data Protection Regulation (GDPR), giving rise to cookie banners.
Unfortunately, the design of these banners often proves problematic, regularly preventing internet users from making a choice aligned with their data-sharing preferences.
To remedy this, the DITP and the CNIL have asked BIT for support in carrying out an online experiment involving 4,000 adults living in France. The aim was to gain a better understanding of Internet users’ data-sharing preferences and to assess the impact of cookie banner design on Internet users’ choices, including the impact of dark patterns (harmful design) and bright patterns (a design that, conversely, encourages Internet users to reflect on their choice).
The results confirm the existence of a paradox where the majority (93%) of Internet users consider the protection of their privacy to be a priority, but only a minority pay attention to how cookies collect their data. These initial results suggest that current cookie banners do not allow users to make an informed choice that respects their preferences.
The results demonstrate the considerable impact of banner design, whether negative or positive, on choice. Banners that make rejection more complicated (because the option to reject is less visible or requires more clicks) encourage many more Internet users to accept. Banners highlighting the option to reject cookies, or even going so far as to point out that acceptance results in data tracking, significantly reduce the acceptance rate.
These results suggest that more attention needs to be paid to the regulation of cookie banners, given that when used wisely, the design can even encourage participants to make a choice that is closer to their data-sharing preferences.
