Recommendations on Load to Discover Task

Hey! The biggest perk here is the modular ETL structure. Unlike the old, rigidly coupled connectors, xFlow lets you run a single extract and route it to multiple loaders at once—like grabbing data from an SFTP and sending it to both Discover and Engage simultaneously. Plus, it brings SSH key encryption support for SFTP/S3 right out of the box and uses a UI that should feel pretty familiar if you've spent time in legacy Qualtrics.

That said, we are still in the early phases. The current goal is achieving "good enough parity" with the old connectors. While the long term strategy is to completely deprecate those legacy connectors in favor of xFlow ETL jobs, there is no official timeline for that deprecation yet. Howeverr, all new features, enhancements, and future Ai developments are happening strictly within xFlow, while the old connectors are essentially on life support for maintenance only.

If you are thinking about jumping onto the early  access/preview for this, just keep a few constraints in mind:

• Your Discover and Engage environments must already be connected.

• Your inbound data cannot require any PCI redaction.

• It currently only supports basic transformations.

• Supported use cases include loading Engage survey, ticket, or IDP data into Discover, or pulling conversational data from SFTP/S3 using the exact same file formats the legacy Discover SFTP  connectors used.

We don't have hard numbers on throughput or latency differences yet, as the documentation focuses purely on capability and security rather than raw speed benchmarks.