Forum Discussion
Groupby, SUM based on interval
Hi all,
Need help to visualize compensation amounts based on fixed values and amount of consumptions
Data
We have a table containing consumtions values (numeric, decimal)
Fixed values table
ConsumptionInterval,compensation
7000-7999,100
8000-8999,200
9000-9999,300
This needs to be "connected" to the dataset that contain the consumption and present a aggregated amount of compensation sum per interval
Ex:
How much compensastion should be payed to accounts in the different intervals
Note that only specific months are in scoop (in ex above december)
Any ideas ?
Br
erik
5 Replies
- DataInsights
Super User
This solution requires a slight modification to your fixed values table (ConsumptionInterval). See below:
There is no relationship between the tables. Create the measure below:
Compensation Total = SUMX ( FactTable, VAR vConsumption = FactTable[Consumption] RETURN CALCULATE ( MAX ( ConsumptionInterval[Compensation] ), vConsumption >= ConsumptionInterval[Consumption Interval Start], vConsumption <= ConsumptionInterval[Consumption Interval End] ) )- eriweste
Helper I
Hi, thanks for the reply and sorry for the delay in response.
The formula works good, but I discovered an additional issue
More info:
The consumtion column get summarized in my table, some "Instnum" consist of multiple rows within the same date range (ex Month = Dec)
This creates the sumx formel to be used twice (2)
This customer hits the Interval range twice
Above screenprint have the consumption summarized (total) and thats what the compensation should be based upon.
ANy further ideas ?
Br
erik
- Ashish_Mathur
Super User
Hi,
Share a dataset and show the expected result.
- AnonymousNot applicable
Hi eriweste ,
Does DataInsights 's reply solve you problem?
If the problem is solved, please mark the helpful reply as the solution, more people will benfit.
If not, please provide the expected result.
I'm wondering if you want the accumulated value, for example if Consumption is 8500, your result is 100+200=300
Best Regards,
Stephen Tao
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.