Event Sourcing vs. CDC for Microservice Data Sync: Pros and Cons?

Hey everyone, I'm working on a new microservices project and we're debating the best approach for keeping data synchronized between services, especially when a service owns a specific bounded context but other services need to react to its data changes. We're looking at two main patterns: Event Sourcing and Change Data Capture (CDC). Event Sourcing seems powerful for auditing and replayability, but it feels like a significant architectural commitment and might introduce complexity for simple CRUD operations. CDC, on the other hand, appears less intrusive, potentially easier to implement with existing databases, but I worry about schema evolution and the potential for a 'data dump' rather than meaningful events. Has anyone implemented either of these extensively in a microservice setup? What were your experiences with developer productivity, operational overhead, and the ability to evolve your system? Are there specific scenarios where one clearly outperforms the other?