Making Grab’s Everyday App Super

Summary

Grab manages an expanding superapp ecosystem comprising ride-hailing, food delivery, payments, and partner content surfaced through the Grab Feed. As content volume grows, the platform risks overwhelming users with irrelevant information. To address this, Grab built a recommendation engine that ranks cards using signals across user profiles, content metadata, and contextual factors such as time and location. The system employs multiple recommendation strategies—including popularity metrics, user favorites, collaborative filtering, habitual patterns, and cross-platform deep embeddings—which are selected or aggregated. Recommendation quality is evaluated via offline metrics like Recall@K and NDCG alongside online engagement experiments.

Takeaways

1. 1

   The recommendation engine relies on three core signal categories: user history across services, item metadata and consumability, and contextual factors like time or GPS coordinates.

2. 2

   Interaction weights must decay over time to prevent stale user preferences from degrading recommendation relevance.

3. 3

   Performance and quality are tracked through offline metrics like Recall@K and NDCG, combined with online metrics such as interaction, clickthrough, and engagement rates evaluated on an experimentation platform.