Pinterest, the popular image-sharing platform, has provided insights into how its algorithm works and has emphasized the risks of relying too heavily on user engagement. In a new document titled ‘Field Guide to Non-Engagement Signals’, Pinterest proposes non-engagement signals as a solution to balance content ranking and improve the user experience.
The platform believes that excessive reliance on user engagement to rank content can result in a negative user experience. It can lead to the surfacing of low-quality or even harmful content that attracts attention but does not provide value to users. To address this issue, Pinterest suggests incorporating non-engagement signals into the algorithm.
Non-engagement signals are generated from two primary sources. The first source is in-app surveys, where users have the opportunity to provide direct feedback about the platform. For example, Pinterest may conduct surveys within the app to gather user insights. The second source is independent assessments of content quality, which are generated through manual labeling.
By understanding how Pinterest’s algorithm works and the importance of non-engagement signals, brands can prioritize the right metrics to secure more visibility for their content. This knowledge allows them to create content that aligns with Pinterest’s values and provides value to users.
To encourage other companies to contribute to “building a more inspired Internet,” Pinterest collaborated with UC Berkeley and the Integrity Institute to create the Field Guide to Non-Engagement Signals. The guide aims to help platforms make informed decisions when utilizing non-engagement signals, rather than dictating what they should do.
The field guide offers several practical applications for product development, including tuning for emotional well-being, using Generative AI to scale content quality signals, and improving user retention. These applications are based on practical industry knowledge and can aid platforms in creating better user experiences over time.
Pinterest’s commitment to inclusivity is also supported by non-engagement signals. When users specify preferences regarding body type, hair pattern, or skin tone in their feed, Pinterest can prioritize relevant content accordingly. This approach ensures that users are exposed to content that resonates with their interests and values.
In a blog post, Leif Sigerson, Pinterest Sr. Data Scientist, and Wendy Matheny, Pinterest Sr. Lead Public Policy Manager, highlighted the importance of non-engagement signals. They stated that optimizing purely for user engagement can surface low-quality or harmful content. By incorporating non-engagement signals, platforms can avoid promoting content that may attract attention but does not provide meaningful value to users.
Overall, Pinterest’s insights into its algorithm and the focus on non-engagement signals demonstrate the platform’s commitment to improving user experiences and providing valuable content. By understanding how the algorithm works and prioritizing the right metrics, brands can enhance their visibility and connect with their target audience effectively. The Field Guide to Non-Engagement Signals offers practical applications for platforms to create a more inspired Internet and ensure that content ranking aligns with user preferences and values.
