I have a sql server connected to a spreadsheet that is used in 10+ dashboards (everything runs on extracts, nothing live). In a new dashboard, I have an additional spreadsheet joined in. I'm wondering what would be more efficient- either adding the join to the original connection that is used by the other dashboards, or keeping this as a separate data source, which means the workbook has an additional data source. Is it more efficient to minimize the number of datasources that a worksheet uses (in which case I will add the additional columns to the connection), or since the one spreadsheet is only used in one dashboard, keep that connection completely separate and only use it for the dashboard it is needed for?
Hi Ben,
Looking at this blog post from a few years ago, Understanding Tableau Data Extracts | Tableau Software, and it mentions that Tableau's data extracts are column-store based as this allows Tableau to only load the columns of data being used in a viz. With this, it seems your best bet would be to just join your additional spreadsheet to your current data source as Tableau will only pull data from those additional columns for vizzes as needed.
Hope this helps!
