Data Source Structure - Best Practice?

[SimonNU]

I understand that rows > columns but at what point should this cease? For example, would it be better to have 1 table of Actual data and 1 table of Budget data or should these be combined?

If anyone has any links to some of the more technical articles on the subject matter then that would be greatly appreciated!

 

[scottsen]

I have never combined budget and actuals, but... I could see it...

Performance is probably my favorite topic, so in the short life of my blog you get quite a few entries there:
How does Power Pivot store and compress data? | Tiny Lizard
Dollars, Cents and #PowerPivot | Tiny Lizard
Power Pivot Performance Gotchas | Tiny Lizard

Other than "lots of columns is bad", the other would be just being aware that "unique values" are bad... so much so, that splitting dollars an cents into 2 columns... helps. (overriding rule #1 )

 

[PentaGalCXO]

Every instance where I work with Budget vs. Actual the data sets are at a different granularity... forcing me to put them in differenct tables with common deminsions... Even when they are the same time granularity (such as monthy Budget and monthly Actuals) I get bit by the fact Budget may be assigned at Territory and Actual is at specific location, or Budget is at Customer rollup and Actual is at a specific customer... Generally better off keeping Budget/Actuals in seperate tables...