E pluribus unum — from many, one. With a slight edit, “from many, one a few,” we have a good starting point for understanding why anyone might want to go from “sprawling dbt project” to the data vault modeling methodology. Because amongst that sprawl (the many) are really only around a dozen-ish overarching business concepts (the few).
There is a lot of good information out there about what data vault is, at a very high level. But not necessarily a glut of documented user experiences and stories from those who have implemented data vaults. Back at the tail end of 2023, I had the opportunity to present my “fifty first data vaults” to the Data Vault User Group. During the session I gave a quick 411 on my setup and then we got into the weeds of what went wrong in each iteration. We ended with why data vault is ultimately worth the heartache.
If you’d like to listen to the recording: