Yeah it was a report designer, we do some analysis in our group now, but the nature of our structure and using Data Warehousing is different, and you can still do raw SQL pulls against it, we just happen to have several BI tools that make it so you don't have to. We can't teach our 5000 users SQL and most of them don't have the time. So we are reliant on doing good ETL and making sure we work hand in hand with the business to ensure we encompass all of their data needs. It is still a tried and true DW model.
A lot of the time our data marts are just massive relational star-schema DBs with tons of data and a lot of ways to dive in to it. Other times it is already very sliced up due to business rules.
I get what you are saying for sure though and I agree, you want to avoid a lot of assumptions. Especially going forward in analytics you don't want a lot of those presumptions. And when I say I have an ETL dev create a view, it is literally just, get me these four tables from DB A, these four from DB B and these 6 from DB C and give me these joins. It is what the guy does all day and he can turn around a MV real quick and then I can go in and pull in that data to my tool and do the actual analysis with no preconceived assumptions.
Also, just in general, I feel like the more and more I get exposed to new concepts and the way other companies operate in Data Science and Analytics, the more I think my current company isn't completely doing it right or completely understanding new concepts. I want to keep building and growing in this, but man it feels like they are missing the boat.
A lot of the time our data marts are just massive relational star-schema DBs with tons of data and a lot of ways to dive in to it. Other times it is already very sliced up due to business rules.
I get what you are saying for sure though and I agree, you want to avoid a lot of assumptions. Especially going forward in analytics you don't want a lot of those presumptions. And when I say I have an ETL dev create a view, it is literally just, get me these four tables from DB A, these four from DB B and these 6 from DB C and give me these joins. It is what the guy does all day and he can turn around a MV real quick and then I can go in and pull in that data to my tool and do the actual analysis with no preconceived assumptions.
Also, just in general, I feel like the more and more I get exposed to new concepts and the way other companies operate in Data Science and Analytics, the more I think my current company isn't completely doing it right or completely understanding new concepts. I want to keep building and growing in this, but man it feels like they are missing the boat.
Last edited: