What is microservices architecture?

The defining property is independent deployability: each service owns its own data and can be released without coordinating with the others. That is genuinely valuable when separate teams would otherwise block each other in the same codebase and release train. The cost is that every call between them is now a network call that can be slow, fail, or succeed twice.

Which means you inherit a distributed system's problems. An operation spanning two services can no longer be one database transaction, so consistency has to be designed rather than assumed. Debugging needs tracing across services instead of a single stack trace. Local development needs several things running at once. Splitting a well-structured monolith later is usually easier than merging services you split too early.