Magic ETL Tiles groups for clearer organization and visibility/observability :D :D :D


Although similar to the "Allow users to specify the color of Magic ETL tiles",

we sometimes (okay, often) have extremely long and complex Magic ETL's that are difficult to comprehend, especially if you're inheriting them or exploring for replication or use. Being able to "group" your tiles in some fashion would greatly enhance our ability to understand what each "branch" or section intends to do (think "remove PII section", "split by product for vertical-specific logic", "add dimensions", etc). If these "groups" had descriptions for the groups or even titles/headers for additional clarity and metadata, that would be a significant boost.

I think pairing this with the color of Magic ETL Tiles would allow lots of customizability for organizations to help ease the burden of trying to understand a complex Magic ETL, reducing time spent on troubleshooting, or even documentation.

here's an example from Count for their metric trees template

and an example of how i document processes/ETL's using Lucid:

18 votes

Scheduled on Roadmap · Last Updated

Thank you for your feedback. We've started UX conversations for cleaner tiles.


  • DomoDork
    DomoDork Contributor

    I 100% agree with this. If we could group tiles together along with colorize tiles, it would make complex ETLs much easier to work with. I would also extend the colorizing tiles idea such that when you colorize tiles and run a preview to see sample data from each tile it also colors the column headers. It would make it much easier to tell where a column came from in your ETL flow.

  • jimsteph
    jimsteph Contributor

    Ten thousand thumbs up for colorizing the column headers to see where the column came from!

  • matthewpinson

    This is sick. Love the bit on your current process as well.

  • erikjamesmason

    @matthewpinson thank you kind stranger whom i've never met before

  • matthewpinson

    Yes, indeed, I definitely agree I am kind, and a stranger.