Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 

Earn the coveted Fabric Analytics Engineer certification. 100% off your exam for a limited time only!

Reply
nhanser
Regular Visitor

DAX Running total by another column

Hello,

 

I need help to create a DAX column/measure that will create a running total by another column.

 

So for example, looking at this table

 Item  |       Date      |  QTY
----------------------------------
    a    | 12/31/2015 |    1
    a    | 12/31/2015 |    -1
    a    | 12/31/2015 |    2
    b    | 12/31/2015 |    10
    b    | 12/31/2015 |    -5
    b    | 12/31/2015 |    1

What I need at the end is to add another field/column that will create a running total, something like this:

 Item  |       Date      |  QTY  | RunTotal
------------------------------------------------
    a    | 12/31/2014  |    1     |     1
    a    | 1/15/2015    |    -1   |      0
    a    |  2/28/2015   |    2     |     2
    b    | 12/31/2015  |    10   |   10
    b    | 4/6/2015      |    -5   |     5
    b    | 5/5/2015      |    1     |    6

I have tried using a calucalte function and filetering on the date, but it only seems to create a running total for everytihng, not filtering by the item.  Similar to this:

 

On Hand = CALCULATE( SUM('Table'[QTY]),
	FILTER(
		ALL('Table'),
		'Table'[Date] <= MAX('Table'[Date])
	)
) 

 

Any ideas would be much appericated.  Thanks for any and all help in advance!

1 ACCEPTED SOLUTION

Hello @nhanser,

 

for me it looks like you want to add a calculated column, but the DAX code you use is for beeing used in a pivot table. 

 

I copied your raw data, but used different dates.

 

raw data

 

The pivot I created looks like this and I hope this is what you want:

 

Pivot

 

The code I used is close to the one you used, but slightly different:

 

RunningTotal :=
CALCULATE (
SUM ( Tabelle1[QTY] );
FILTER ( ALL ( Tabelle1 ); Tabelle1[Date] <= MAX ( Tabelle1[Date] ) );
VALUES ( Tabelle1[Item] )
)

 

The key here is the VALUES() which puts the Items into the filter context of the CALCULATE()-statement. Otherwise you would have running totals on the dates, but it would be the same for all your Items. 

 

Hope that helps a bit.

 

Greets,

Lars

View solution in original post

15 REPLIES 15

Hello @nhanser,

 

for me it looks like you want to add a calculated column, but the DAX code you use is for beeing used in a pivot table. 

 

I copied your raw data, but used different dates.

 

raw data

 

The pivot I created looks like this and I hope this is what you want:

 

Pivot

 

The code I used is close to the one you used, but slightly different:

 

RunningTotal :=
CALCULATE (
SUM ( Tabelle1[QTY] );
FILTER ( ALL ( Tabelle1 ); Tabelle1[Date] <= MAX ( Tabelle1[Date] ) );
VALUES ( Tabelle1[Item] )
)

 

The key here is the VALUES() which puts the Items into the filter context of the CALCULATE()-statement. Otherwise you would have running totals on the dates, but it would be the same for all your Items. 

 

Hope that helps a bit.

 

Greets,

Lars

@LarsSchreiber Is it possible to create the 70th percentile on the results per each Item?

 

Thanks,

Zaid

Hi @LarsSchreiber,

your solution didn't workout in my scenario..!!

 

DimItem(ItemKey,ItemNumber,CompanyKey) (One ItemNumber can have Multiple ItemKey because of CompanyKey)

FactRevenue(ItemKey,CompanyKey,CustomerKey,BrandKey,AccountingDate,Revenue) (Fact is day level)

Fact and Dim Item Joined on ItemKey.

when i try to put ItemNumber from DimItem and Revenue from FactRevenue in a table visual with following Measures.

 

TotalRevenue = CALCULATE(SUM(FactRevenue[Revenue]),ALL(DimItem[ItemNumber]))

RContr =  DIVIDE(SUM(FactRevenue[Revenue]),[TotalRevenue],0)

RT = CALCULATE([RContr],FILTER(ALL(FactRevenue),FactRevenue[AccountingDate] <= MAX(FactRevenue[AccountingDate])),VALUES(DimItem[ItemNumber]))

 

I am not getting good results. Please consider the output:

5.99+5.04 = 11.03 or something.

Capture1.PNG

@LarsSchreiber I need your help. I am looking for same solution but somehow its not working for me. 
running total.JPG

RunningTotal =
CALCULATE (
SUM ( TabelForecast[Forcast] );
FILTER ( ALL ( TabelForecast ); TabelForecast[MonthYear] <= MAX ( TabelForecast[MonthYear] ) );
VALUES ( TabelForecast[M#] )
)

 

 

I have tried your formula in power BI but its not giving me running total but total for all the rows 😞 Any suggestions?

 

 

Hi @FawadRehman,

 

sorry, I haven't been here for a while. Could you solve your problem already?!

 

regards,

Lars

 

 

This is a great solution, thank you. Worth noting you can stack multiple variables in the value cluase to allow aggregation across multiple / different dimensions, (similar to partition by in sql I guess).

 

However, I have run into a slight issue. If you have a value or combination of values that are empty i.e. there is no row in your dataset for that value or combination of values, the running total becomes incorrect until the next time a row matching those criteria exists, at which point it 'catches up'.

 

For example, if you are creating a running total of sales by month and team across a year, using a data set like:

 

Month Team Sales

Jan      A       100

Jan      B       200

Feb     A       150

Feb     B       210

Mar     A       300

Apr     A       120

Apr     B       400

 

In March, if Team B has a month where they have no sales (and therefore no row in the dataset), your running total for the business as a whole will ignore all Team B's previous data in the calculation of a running total, returning 550 as the total, rather than 690.

 

I'm yet to find a fix for this (despite habving tried every combination of isempty i can think of). If anyone has any ideas I'd love to hear them


Robbie

 

The following month if Team B has

@robbie337 I am having the exact issue as you. Have you found a solution yet? 

Hi @robbie337 did you find a solution to your problem?

 

I have a similar problem, since I want to display my Running Total Measure in a Line Chart and I need to to filter on different categories in a slicer. 

 

If I include my VALUES[Category column] in the Running Total Measure all is fine as long as I ALWAYS have one of the Categories selected in my slicer. If I deselect all in the Category Slicer the Line Chart will jump up and down and not continue to increase as intended.

 

To fix this I can just leave out the VALUES([Category column] of my Running Total Measure, but then my Running Total would just show an aggregation across all categories and not calculate PER category as intended.

 

 

...This is such an easy task in Excel Pivot Tables where Running Total Calculation is a standard option - why is this still missing in PowerBI.  

Hi @Mardin,

Did you get any solution for this issue?

Hi @sridhark Yes, since Quick Measures was introduced i PowerBI now I actually just created a Quick Measure for my Values column and selected Running Total by Date (remember to have a seperate Data Dimension table and use this with your Running Total measure).

It now works with 3 different dimension slicers as I needed.

 

@LarsSchreiber - I was looking for a similar solution as @nhanser. Your solution worked perfectly!

 

The key, as you mentioned, is the VALUES() function to put the running total calculation in context of the field needing to be grouped.

 

Thank you for the very useful response!

@jmalone,

 

I am happy, when it helped 🙂

 

Regards,

Lars

 
PowerBIGuy
Responsive Resident
Responsive Resident

Have you tried:

 

On Hand = CALCULATE( SUM('Table'[QTY]),
	FILTER(
		ALLSelected('Table'),
		'Table'[Date] <= MAX('Table'[Date])
	)
) 
Business Intelligence Architect / Consultant

Helpful resources

Announcements
April AMA free

Microsoft Fabric AMA Livestream

Join us Tuesday, April 09, 9:00 – 10:00 AM PST for a live, expert-led Q&A session on all things Microsoft Fabric!

March Fabric Community Update

Fabric Community Update - March 2024

Find out what's new and trending in the Fabric Community.