Month by month variance from previous month

[colin_dickson]

Hi,

I'm fairly new to Power BI.

I have a simple matrix table which has types of service on the rows, months as columns and the some of each as the values, as follows:

Service	Jan	Feb	Mar
Service 1	500	800	300
Service 2	600	900	400
Service 3	700	1,000	500

What I want to be able to to do is convert that so I can see the variance from the previous month i.e. Service 1 from Jan to Feb would be a positive variance of 300 and so on.

I've tried all sorts but nothing seems to work.

Any advice would be much appreciated.

Thanks

Colin

[colin_dickson]

I fixed that...

It is now only showing one column with variances I don't recognise... 😞

Regards

Colin

[colin_dickson]

That makes sense...

I'm getting the following error when I add that measure?

OrderBy column 'Month No.' could not be found in OFFSET's Relation parameter.

Regards

Colin

[Jihwan_Kim]

Hi,

I am not sure how your datamodel looks like, but please check the below picture and the attached pbix file.

expected result measure: =
VAR _currentvalue =
    SUM ( Data[Value] )
VAR _prevmonthvalue =
    CALCULATE (
        SUM ( Data[Value] ),
        OFFSET ( -1, ALL ( 'Month' ), ORDERBY ( 'Month'[Month No], ASC ) )
    )
RETURN
    IF (
        NOT ISBLANK ( _prevmonthvalue ) && HASONEVALUE ( Service[Service] ),
        _currentvalue - _prevmonthvalue
    )

[colin_dickson]

Thanks for this, i will give it ago.

I haven't got a data model as you sugest, all of my data work is done in the transform section then i'm using the final appended and  merged data as my source data.

Why do we need a month number, I assume I need to add this to my data?

Regards

Colin

[Jihwan_Kim]

Hi,

Thank you for your message.

The reason why we need a month number column is that,

- Jan, Feb, and Mar has to be sorted by month number column, not by alphabetical order.

- we need to let Power BI know that "Feb" minus one is "Jan".

The month number column helps these kinds of sort-order-arrangement and calculations.

I hope this helps.

Thanks.