Calculated Table for Month over Month Changes

[xzfujc]

Hello,

I have an archive table that looks like the following:

Gate	Emp. ID	Count	Month	Year
IA	a123	351	9	2018
IRA	a123	220	9	2018
NEW	a123	26	9	2018
PSA	a123	108	9	2018
IRA	b456	39	9	2018
NEW	b456	151	9	2018
PSA	b456	55	9	2018
IA	a123	453	9	2018
IRA	a123	125	10	2018
NEW	a123	35	10	2018
PSA	a123	99	10	2018
IA	b456	20	10	2018
IRA	b456	9	10	2018
NEW	b456	109	10	2018
PSA	b456	66	10	2018

I want to create a calculated table to show the Month over Month Change in the count of "Gates" from month 9 to 10 by Employee ID (Emp. ID).  So I want a table that will look like the following: 

Gate	Emp. ID	Month Over Month Change
IA	a123	102
IRA	a123	-95
NEW	a123	9
PSA	a123	-9
IA	b456	20
IRA	b456	-30
NEW	b456	-42
PSA	b456	11

This would allow me to show how things are moving from one gate to another each month for each employee and through the entire department. How can I code this in DAX so that the count from the Maximum Month in the table subracts the count from the previous month for each gate and employee combination? 

I would need code to take into consideration how employee b456 has no IA gate in month 9 but has a count of 20 in the IA gate by month 10. Also I would need the code to take into consideration if the month is 1 then to subtract the count from month 12 of the previous year. 

Any advice would be appreciated! 

[dedelman_clng · Accepted solution]

Hi @xzfujc

The best way is to think of measures instead of tables/columns.  You can then calculate the number of gates for the given month, the number of gates for the previous month, and then the difference.

You will need a date column to key off of, but since you have Month and Year, that's easy enough to create.

//Calculated Column
Dt = DATE(Year, Month, 1)

//Measures
Total Gates = CALCULATE(SUM(Count))

LM Total Gates = CALCULATE([Total Gates], PREVIOUSMONTH(Dt))

Variance = [Total Gates] - [LM Total Gates]

This will allow you to create a table visual, and if you want to look at it by just Gate instead of by Gate and Emp ID, the calculation will take care of itself.

Hope this helps

David

[xzfujc]

I can't explain it, but adding the .[Date] to the Dt field solved my problem. After adding the .[Date], the LM value was calculated correctly. 

LM Total Gates = CALCULATE([Total Gates], PREVIOUSMONTH('Gates Count Archive'[Dt].[Date]))

Thanks for all your help!

[xzfujc]

I think your solution will accomplish what I need. 

I am running into an issue with 

LM Total Gates = CALCULATE([Total Gates], PREVIOUSMONTH(Dt))

It returns a 0. So my total variance is now equal to my total gates. Working on resolving. Is there a way to say something like the following?

Total Gates = CALCULATE(SUM(Count), CurrentMonth(Dt))

[dedelman_clng]

Change the variance calculation to

Variance = IF(ISBLANK([LM Total Gates]), 0, [Total Gates] - [LM Total Gates])

(or use BLANK() instead of 0 if you want to show no value in Variance for the first month)

Hope this helps

David

[xzfujc]

Really appreciate all the help, and I do believe the solution you provided is the solution I need. I am having trouble fully implementing it though, for some reason I can not get the LM Gates Total to sum anything which gives me a variance of 0. 

Is there any reason you could think of why the the LM Total Gates returns null?

[v-cherch-msft]

Hi @xzfujc

You may drag the 'Dt' column to the table visual instead of 'Month' column.

LM Total Gates = CALCULATE([Total Gates], PREVIOUSMONTH(Dt))

Regards,

Cherie

[dedelman_clng]

Also, you should have a dedicated CALENDAR table that has a relationship to Dt on your data table in order to fully utilize time intelligence like PREVIOUSMONTH()