Cumulative Count by Date help

[powerbiuser101]

Hi,
I think I'm very close but I can't figure out the last piece of the puzzle. The solution here might be close but I need : https://community.powerbi.com/t5/Desktop/Cumulative-Count-by-Date/m-p/368918#M167100

Here's a sample of my data:

In Date	ID	Completed	Out Date
11.05.2018	A1	yes	11.06.2018
18.05.2018	B2	no
20.05.2018	C3	no
06.06.2018	L13	no
17.06.2018	D4	no
18.06.2018	E5	yes	20.06.2018
18.06.2018	F6	no
19.06.2018	G8	yes	19.06.2018
24.06.2018	H9	yes	26.06.2018
25.06.2018	I10	yes	26.06.2018
25.06.2018	J11	no
26.06.2018	K12	no
	M14	yes	03.06.2018

What I need to do is:
* Counts of [Complete] = "no" on a day/monthly basis (preferably weekly but day/month was more priority)

This would mean, today the 27th June, a line chart should show a count of 7 no's. I want to show a line chart of outstanding incompleted IDs going back 4-5 months.

Any assistance appreciated!

[Seward12533]

Unless I'm missing something I think all you need is to do define a measure for the count of incomplete

Count InComplete = CALCULATE(COUNTROWS('Count Complete'),'Count Complete'[Completed]="No")

And then write a measure for Cumulative count of that measure similar to the one you posted.  Assuming you have a date table linked you can then plot this over any period you want Day,Week, Month, Year, Quarter, etc...

You may also want to look at The TOTALMTD function

[powerbiuser101]

Hi @Seward12533,

Unfortunately no success. When graphed, for example, expectation would be that on June 17th, the value of the red line would be at 4. 

I'll look into TOTALMTD. Thanks

[Seward12533]

You need to build a mesure to calcate cumulative Count of Incomplete either by using the built in TOTALMTD, TOTALYTD or building your own similar to the post you linked to in your original question. Someting like

Cumulative Count of Inc = IF(MIN(Calendar[Calendar_Date])<=NOW(),CALCULATE([Count Inc],FILTER(ALL(Calendar),Calendar[Calendar_Date]<=MAX(Calendar[Calendar_Date]))))

The IF blocks this from calculating for dates in the future assuming your date table does as well.