Paris, 31 May 2022 – Commanders Act, the publisher of a cookieless marketing platform, has unveiled the fifth edition of its Privacy Barometer, which provides insights into the performance levels of the mechanisms that its customers have implemented to collect consent in accordance with the GDPR. The findings of the 2022 edition of the Barometer reveal that the opt-in share, i.e. the number of consents vs the number of users who expressed a choice, was down slightly on 2021, but still high (average of 74% for desktop users). Although the vast majority of websites in France now use a consent banner in accordance with the data protection authority’s recommendations, an examination of their banner’s performance reveals a number of disparities.
To produce this barometer, Commanders Act analysed 285 consent banners on websites representing the main sectors of activity in the French market between 1 and 31 January 2022.
Since 31 March 2021, France’s data protection authority (CNIL) has required companies that are active on the internet and mobile devices to comply with the guidelines governing consent collection and cookie management. In particular, these directives impose the use of an explicit mechanism for collecting consent, featuring at least two separate action buttons of the same level (“Accept” and “Refuse”).
One year after the new directives were implemented, companies have widely rolled out their explicit consent banners. Most have gone for a pop-in banner, which requires users to make a choice before they can continue browsing on the website. Web visitors are well used to seeing this type of mechanism, meaning that they are more adept at understanding and dealing with banners than before.
“The market is tending to align with the same banner model and offering few innovations,” explains Michael Froment, CEO of Commanders Act. “As with any practice entering the mainstream, it is accompanied by a slight fall in performance with an average opt-in share of 74% for desktops compared to 81% in 2021, and an average consent rate of 45% for desktops compared to 55% in 2021.“
However, not all mechanisms offer the same performance: “We’ve seen that some characteristics relating to the design, look & feel and user experience have a tremendous influence on the opt-in rate,” stresses Michael Froment at Commanders Act. “The best practices that were already effective in 2021 are still effective today, and new models deserve to be explored and tested.“
The best opt-in rates are achieved with a banner model that includes a “Continue without accepting” button or link, and an “Accept” button. The three-button model (“Accept”, “Refuse” and “Configure”) is very effective for websites with a premium brand image. Choosing an appropriate colour for the “Accept” button still plays a major role in gaining the user’s consent. “For example, using red for the “Accept” button, even if red is part of the company’s style guide, will clearly cause the consent rate to plummet,” explains Michael Froment.
It is vitally important to use a clear scenario and avoid forcing users to click several times if they do not wish to provide their consent. This practice is not only skating on thin ice when it comes to complying with CNIL’s recommendations, but it also leads to a catastrophic bounce rate that can reach 70% in the worst cases. Another pitfall is requiring users to give their consent whenever they change domain, which often happens with multi-domain websites. “Only ask the question once! Repeatedly asking for consent is guaranteed to aggravate users and lose their consent,” advises Michael Froment.
Finally, even though companies are tending to use the same type of model, some websites are taking a more original approach with creative banners that are ultimately more engaging. “Although our recommendation is to prioritise sober and classic designs, some companies have managed to strike the right balance between the legal obligation for collecting consent and their brand’s defining features and personality,” explains Michael Froment. “It’s an extremely interesting approach to the subject and it could inject new driving force into the market and help companies achieve great performance, but they really must carry out A/B testing to make sure that their banners are effective. Unless they can perform that type of test, the conventional approach is currently the best choice.“