Many adults frequently encounter content online that they consider to be harmful, yet very few make use of existing content control tools and safety features. This raises an important question: how could we unlock choice and give users more control over the content they see?
This research examines how the presentation of sensitive content settings influence user choices. We conducted a large-scale randomised controlled trial (‘RCT’) with a nationally representative sample of over 3,650 UK adults to test the impact of different choice architectures – such as wording, layout, and the level of granularity – on the decisions users made regarding their sensitive content controls.
The results provide strong evidence that the granular choice intervention was the most effective in promoting safer online behaviour, with more than seven in ten participants using content controls to reduce their exposure to sensitive content. In contrast, changes to wording had no impact, while visual cues increased uptake of safer settings but were sometimes perceived as less trustworthy. The effectiveness of content controls can vary across different user groups, with age, gender, and online habits shaping safety choices.
So, what does this tell us? How we present online safety information matters, and it is important for platforms to consider how best to design effective safety tools that help unlock choice and empower users to take more control over the content they see online.
For more details about the key findings, see our Insights Paper
For more technical details about the trial design and data analysis, see Technical Report