Blanks vs Zero's when utilizing an Excel Doc as a Datasource

[chris1234512]

I have an excel doc as a data source for a power bi model. The doc utilizes multiple division formulas so in some circumstances I'm receiving a #DIV/O! error. I generally fix this error on the excel side, by utilizing IFERROR and making the result a blank. From a Power BI standpoint when I use this sheet as a source doc, is it more efficient to make the results zeros or blanks?

[kpost · Accepted solution]

as @sayaliredij said, you can fix them on the Power BI side instead using "DIVIDE" like this:

New Column = DIVIDE([Value A], [Value B], <put 0 or BLANK() here depending on what you want in error case>)

In my opinion, though, as far as Power BI is concerned, whether you do it in Excel or Power BI, best practice is to fill with zeros.  It is more likely to result in desired behavior in the visuals you end up creating using that data set.

[chris1234512]

Sounds good. I'll roll with the zeros then. Appreciate it!

[sayaliredij]

HI @chris1234512 ,

You can use DIVIDE in DAX formule instead of using '/' . it will handle the error like excel

https://learn.microsoft.com/en-us/dax/best-practices/dax-divide-function-operator

Thanks,

Sayali

[chris1234512]

For my purposes, I'd like to keep the division in excel. Any tips if that's the case?