Daily totals from cumulative total

[dpearson123]

Hello all

I am trying to caculate the two empty colulmns in the below: MonoPagsSinceLastReading &  ColourPagsSinceLastReading. The data set is a running totals of usage readings from a fleet of printers and I need to caluclate the the daily readings by serial number.  What would the best way to do this? 

Would it be to create a measure do it for the whole fleet and then it can be filtered? Or create the colmn. It would need to be something like PagesSinceLastReading = current reading - previous reading if serial number is the serial number is the same. 

I am using the below at the moment but it doesn't take into account mutiple readings on the same day and the initial reading. So if a printer first comes online and already has 100,000 prints on it, it'll show 100,000 as it's daily usage. 

dailyMono2 =

VAR CurDay = Counters[ReadingDate]

VAR T1 = FILTER(Counters,'Counters'[SerialNumber]=EARLIER(Counters[SerialNumber]))

VAR PreDay = MAXX(FILTER(T1,Counters[ReadingDate]<CurDay),Counters[ReadingDate])

RETURN

Counters[Mono] - MAXX(FILTER(T1,Counters[ReadingDate]=PreDay),Counters[Mono])

Any help would be great please 

Thanks

David 

[amitchandak · Accepted solution]

@dpearson123 , if you have multiple reading a day and you do not have date-time field then add an index field

https://stackoverflow.com/questions/45715963/creating-an-index-column-for-power-bi

and use that

dailyMono2 =
VAR CurDay = Counters[index]
VAR T1 = FILTER(Counters,'Counters'[SerialNumber]=EARLIER(Counters[SerialNumber]))
VAR PreDay = MAXX(FILTER(T1,Counters[index]<CurDay),Counters[index])
RETURN
Counters[Mono] - MAXX(FILTER(T1,Counters[index]=PreDay),Counters[Mono])