PostgreSQL: Is there any way to further optimise ASYNC replication towards less data loss during failover?

We are looking to see if there any further optimisations that can be done in PostgreSQL(13) ASYNC replication to make data loss near to zero as possible when a failover is triggered during disaster. We don't want change to synchronous replication as with the testing the TPS is really low comparing with ASYNC method. We see an average write lag of 200-300KBs and at peak traffic up to 20MB. Is there any optimisations can do on WAL level? Any other area to tune? **UPDATE** Could wal_compression help optimise this?