Case Study: Second Dinner achieves real-time in-game personalization with Databricks

Reaching new levels of in-game engagement during play

Second Dinner, the studio behind Marvel Snap, needed a faster way to personalize its in-game shop and engagement experiences for players worldwide. Working with Databricks and the Databricks Data Intelligence Platform, the small team moved away from manual, general offers toward real-time, one-to-one recommendations based on player behavior.

With Databricks, Second Dinner centralized data workflows, trained and retrained machine learning models, and used time-sensitive triggers to deliver personalized card bundles and next-best actions. The impact was rapid: the team set up a personalized system in less than a week, updated it daily, and increased experimentation from about 10 times a year to weekly while improving player engagement and purchase relevance.

Second Dinner

Ted Li

Associate Director of AI