first, do you need all the 50 columns?
Is necessary analyze the 30kk rows? Did you use any filter?
Great point. I think I can go back and filter more columns, but I will need all 30,000,000 rows in order to analyze at least 2 years of past sales data.
I'll let you know if this works.
Thanks for your insight!
Yes, I did try connecting our DWH tables to Tableau Desktop directly, and I think it was in fact little bit faster to load the data this way. But I guess I am confusing a bit because my understanding of Tableau Prep is make the data lighter by filtering and joining tables before data is loaded in Tableau Desktop/Online.
The reason for this suggestion was that it would enable you to use incremental refreshes of your data rather than trying to refresh the entire data set each time.
Thanks for your advise. After couple of trials and errors, I ended finding that it was so much faster to load the data directly to the Tableau Desktop by doing just followings.
1) Hide all unnecessarily columns. (I started off with 50 columns which i thought i could use in my analysis at first, but I ended up with only 20 columns that I actually need)
2) Use data filter in Tableau Desktop, rather than Tableau Prep, and choose Incremental Refreshes option. (When you connect Tableau to Salesforce, you wont be able to use "Live" anyway)
I guess the biggest take away from this was not to use Tableau Prep;-( I'm pretty sure there are needs for it but it wasnt for me.
So, thanks again!
Tableau Prep is a really good tool BTW. I just think that, for this use case, there was a more appropriate solution. There are a number of other use cases where Prep will be invaluable.