Formula for calculating rolling 365

[Anonymous]

Hi,

I have a measure, NPS = ([% of Promoters]*100) - ([% of Detractors]*100) that I am trying to create a rolling 365 day calculation for (NOT A SUM). Basically, for every day I would need to calculate this NPS measurement for the past 365 days. 

I'm sure this is really easy but for some reason I can't get it. 

Any help would be appreciated. 

Thanks

[v-jiascu-msft]

Hi @Anonymous ,

Can you share a dummy sample in Text model that we can copy or download?

If the rolling isn't a SUM, what should it be?

It could be like below. 

Measure =
CALCULATE (
    SUMX ( 'table', [NPS] ),
    DATESINPERIOD ( 'calendar'[date], MIN ( 'calendar'[date] ), -1, YEAR )
)

Best Regards,

[Anonymous]

Hi,

So I have these scores over several years from customers. On my line graph/visual, for each day I want to see the NPS calculation for the past 365 days. Ex: If I had a line graph, on January 1, 2018 I would want to see the NPS for the past 365 days. The scores would be from the past 365 days rather than just January 1, 2018 scores.

Customer	Score	Date	Score (Group)
adam	3	10/1/2017	Detractor (0-6)
bob	8	1/1/2018	Neutral (7-8)
dylan	5	2/5/2018	Detractor (0-6)
ed	7	6/8/2018	Neutral (7-8)
frank	9	9/9/2018	Promoter (9-10)
Net Promoter Score NPS = ( % promoters - % detractors ) x 100

[v-jiascu-msft]

Hi @Anonymous ,

What should the result be from your sample here? If you'd like to calculate the values in a period, maybe you can try datesinperiod-function-dax.

Best Regards,