Increasing Watch Time by 131%

Supergreat is transforming the way people discover beauty products. Through user-generated video content, users post their honest reviews, tutorials, and product unboxing videos.

Supergreat faced significant challenges with their previous recommendation system developed with AWS Personalize. While AWS Personalize offers many benefits, it falls short in it's real-time ranking capability. In the beauty industry, it's imperative to stay up-to-date with the latest trends. AWS Personalize was only able incorporate user feedback every 2 hours and re-train models every 7 days, a lag which often resulted in out-of-date or irrelevant recommendations.

This lag contributed to a feeling that the recommendations were 'too generic and not personalized and diverse enough'. After running internal tests, it became clear that AWS Personalize wasn't delivering the increase of engagement expected. Recognizing this shortfall, Supergreat embarked on a mission to improve personalization for their users.

Supergreat rigorously evaluated multiple vendors and considered building their own in-house solution. Ultimately, they chose Shaped, attracted by its team's deep social media expertise, real-time capability, the flexibility of its customizable APIs, and low maintenance required.

Shaped created bespoke ranking models that enhanced personalization and diversified recommendations shown to users. Regarding diversification, remarkably the number of distinct videos watched in the Supergreat library quadrupled. Leveraging Shaped's real-time capabilities, the Supergreat app now seamlessly adapts and personalizes to users' actions in seconds.

Implementing Shaped resulted in a significant boost in engagement. Users are spending more time on the platform, engaging more deeply with the content, and discovering products that are more aligned with their beauty preferences. Furthermore the integration with Shaped was easy, only taking a few days.