Hi everyone! We’re looking to migrate a music / sfx listing and sorting system over to Typesense and could use some guidance on feasibility and best practices. Our requirements are a bit complex: • We need context-specific relevance for different categories (style, tag, artist), ranking tracks by engagement data (plays/downloads) within each context. • To prevent stagnation, we’d like to dynamically factor in recent trends, so newer or trending tracks aren’t “locked out” by historically popular ones. • We’re also considering a “sets” approach where top spots rotate between proven, trending, and new tracks to keep the catalog fresh and engaging for different user types. • Additionally, we’d like to adjust relevance based on user type (e.g., free vs. premium users) and prioritise certain tracks that drive conversions for free users. Does anyone have experience with implementing similar ranking logic in Typesense or advice on structuring this system? We’re particularly interested in whether Typesense’s built-in features (e.g., weighted attributes, faceting) could handle this kind of dynamic ranking and any tips on handling engagement data (e.g., monthly/quarterly segments). We've got a sort-of working prototype but we're especially cautious of doing anything overly computationally heavy / slow and losing out on a lot of the benefits Typesense gives us. Thanks so much for any insights or suggestions!

For 1) 2) and 3) the key is to come up with a single score that encapsulates your business logic and then use one of these mechanisms to combine text relevance with your popularity score: https://typesense.org/docs/guide/ranking-and-relevance.html#ranking-based-on-relevance-and-popularity https://typesense.org/docs/guide/ranking-and-relevance.html#ranking-based-on-relevance-and-recency For 4) here's an article on the topic: https://typesense.org/docs/guide/personalization.html

Thanks @Jason Bosco this is very insightful!