Does a Firebase subcollection's indexed field have its own index and independent write limit?

Firebase automatically creates single‐field indexes on a per‑collection basis. A subcollection—like your Reviews under each User (i.e. User/{documentId}/Review/)—has its own set of indexes. This means that the sequential write limit of 500 writes per second applies individually to each subcollection.

Key Points

• Per‑collection indexing:
  Each subcollection (even though they share the same name across users) is indexed separately. The index entries for a field like createdAt are scoped to that particular subcollection.

• Write limits:
  Since the limit of 500 writes per second applies within a single collection (or subcollection), if each User’s Review subcollection is receiving writes, each one has its own limit independent of the others.

• Path in index storage:
  Under the hood, the full path (User/{documentId}/Review/) is part of the index key information. This means the index entries already account for which User the Review belongs to. There is no need to create a new indexed field such as userDocumentId+createdAt because the path is used when building the index.

Conclusion

If you are concerned about write throughput because of sequential values in the createdAt field, note that each User’s Review subcollection is handled on its own by Firestore. The limitation is applied per subcollection, and the document path is implicitly included in the index. Therefore, prefixing the documentId to the timestamp would not provide additional benefit in this case.