@wleftwich, yes, you still should be able to recover bad data. However, the workflow might be different from what you used to do. That depends on the enrichment version you currently use. There are 2 workflows for streamed data:
- batch pipeline recovery (Hadoop job): Running Hadoop Event Recovery with Dataflow Runner [tutorial]
- [stream] old bad format (Spark job): https://docs.snowplowanalytics.com/docs/snowplow-event-recovery/older-versions/0-1-x/
- [stream] new bad format: https://docs.snowplowanalytics.com/docs/snowplow-event-recovery/
New bad format is discussed here. It was introduced with R118 release.