Hi All,
I've browsed around trying to find an answer to this and have found answers which I've been able to recreate and get working (situationally) for Sumif (singular condition). I need a sumif with two variable conditions rather than fixed.
Basically the two tables I have are like this. 1 shows new lines launched, with a launch date, the other shows sales. I need to measure the impact which sending out free samples has on the first three months of sales - should be easy :P
New Lines:
[Product ID] [Samples sent?] [Launch Month] [Second Month] [Third Month] [Sales in First Month] [Sales in Second Month] [Sales in Third Month]
eg 1234....... yes................... Jan 2015.......... Feb 2015........ Mar 2015.......... SUMIFS()............... SUMIFS().................... SUMIFS()
eg 2345....... No.................... Apr 2015.......... May 2015....... Jun 2015........... SUMIFS()............... SUMIFS().................... SUMIFS()
Sales Table
[Product ID] [Invoice Month] [Sales Value]
eg 1234......... Mar 2015......... 200
Now for normal excel I would do the equivalent of Sumifs(Sales.[Sales Value],Sales.[Invoice Month],[Launch Month],Sales.[Product ID],[Product ID])
I've tried playing around with filters but they don't seem to like having a variable condition, I've also tried playing around with the calculate function - but that doesn't seem to like having multiple filters... I'm kind of going round in circles here and can't find the answers I need on the forums here.
The final step will be easy, build a pivot table to compare first 3 months of sales for all products where samples were sent, vs all products where they weren't
I can do that!!
Thanks in advance for your help. I love these forums and really enjoy learning more tips and tricks
I've browsed around trying to find an answer to this and have found answers which I've been able to recreate and get working (situationally) for Sumif (singular condition). I need a sumif with two variable conditions rather than fixed.
Basically the two tables I have are like this. 1 shows new lines launched, with a launch date, the other shows sales. I need to measure the impact which sending out free samples has on the first three months of sales - should be easy :P
New Lines:
[Product ID] [Samples sent?] [Launch Month] [Second Month] [Third Month] [Sales in First Month] [Sales in Second Month] [Sales in Third Month]
eg 1234....... yes................... Jan 2015.......... Feb 2015........ Mar 2015.......... SUMIFS()............... SUMIFS().................... SUMIFS()
eg 2345....... No.................... Apr 2015.......... May 2015....... Jun 2015........... SUMIFS()............... SUMIFS().................... SUMIFS()
Sales Table
[Product ID] [Invoice Month] [Sales Value]
eg 1234......... Mar 2015......... 200
Now for normal excel I would do the equivalent of Sumifs(Sales.[Sales Value],Sales.[Invoice Month],[Launch Month],Sales.[Product ID],[Product ID])
I've tried playing around with filters but they don't seem to like having a variable condition, I've also tried playing around with the calculate function - but that doesn't seem to like having multiple filters... I'm kind of going round in circles here and can't find the answers I need on the forums here.
The final step will be easy, build a pivot table to compare first 3 months of sales for all products where samples were sent, vs all products where they weren't

Thanks in advance for your help. I love these forums and really enjoy learning more tips and tricks
Last edited: