This reminds me of an interesting project when I was still a consultant. Our partner (audit + consultancy firm) sold the idea of a magical "analytical base tabl

This reminds me of an interesting project when I was still a consultant. Our partner (audit + consultancy firm) sold the idea of a magical "analytical base table" which would consolidate the client's data in a single place for quicker reporting and analysis. Of course all of this was new to me (the ABT thing really obscured the problem). I was going for quality review with another partner at the firm whose bread & butter are data management. Let's say it was a very harsh review. At the end of the review, he recommended that I learn data modelling properly. He gave me a copy of his Kimball book for the December holidays. I went through and completed it, and on the other side of that we were able to build a coherent dimensional data model to address the client's problem efficiently. That knowledge makes me agree with you, merely knowing SQL or your way around a database is often not enough, and knowing some theory about data modelling (regardless of which type you prefer) is a key element in a data engineer's toolkit.