Blog

Pinterest recently spotlighted the "non-engagement signals" used in its algorithm, providing a unique peak into the way Pinterest decides which posts to show users. They are sharing the tips in the hopes that other social companies will adopt the same changes in order to improve user well-being.

Pinterest explains that unlike traditional engagement metrics such as likes and comments, non-engagement signals attempt to look deeper by sourcing user preferences from sources such as:

• In-app surveys.

• Manual assessments of content quality.

• Personalization, such as when a user tells Pinterest the body type, hair pattern, or skin tone they want to prioritize in their feed.

• Optimizing for well-being. Specifically, they aim to “zero-in on specific contributors to well-being” rather than trying to “optimize users’ general well-being”

• Putting more ranking-weight on quality metrics, and using Gen AI to scale quality signals.

Image Source: The Field Guide to Non-Engagement Signals

The approach stems from the realization that default engagement metrics can sometimes spotlight low-quality or even harmful content.

The move to highlight non-engagement signals is meant to align with Pinterest's ethos of inclusivity and quality. The Pinterest Engineering blog explains that the transparency of how this works is part of the company’s Inspired Internet Pledge:

“As the founding signatory of the Inspired Internet Pledge, Pinterest is committed to “sharing best practices, key research findings, and creative solutions across the industry to make the internet a healthier place for everyone.” ”

— Pinterest Engineering Team's Field Guide to Non-Engagement Signals

Pinterest, in collaboration with UC Berkeley and the Integrity Institute, crafted the "Field Guide to Non-Engagement Signals” to publicize their efforts. The guide provides unique insights for other companies working on fine-tuning their own algorithms. It recommends steps such as prioritizing emotional well-being and leveraging Generative AI for scaling content quality signals.

Prioritizing feeds for only engagement often leads to incentivizing click-bait as well as highlighting negative/extreme content. As the Pinterest CEO, Bill Ready, puts it, content ranking can surface the “car crash we can’t look away from.” Prioritizing non-engagement signals is a way social platforms can try to address this harm.

Andrew Hutchinson at Social Media Today summarizes all of this well:

“The bottom line is that focusing on engagement alone will drive more clicks, but the incentives that this drives will also, eventually, degrade a platform significantly.”

Why it Matters

Gaining more understanding of how the black box of an algorithm actually works can help us, as social media marketers, better prioritize our own work to improve the reach of our content.

Thinking more broadly, this news matters because Pinterest’s approach to improving feed algorithms may influence the rest of the social media industry. If it’s successful, this will lead to more wellbeing in society—and happier users on social media.

You may also consider this call to action for marketers to prioritize quality, inclusivity, and user well-being in your own content strategies—thereby fostering a more engaging and positive experience for users across the board​.