Comparing different schedules using a generic month - RankX missing months? Better approach?

[jameschung]

Hi all,

I am trying to compare multiple projects to see when (and how frequently) certain items are occurring during the course of a project.

One project may have started in 2017, another may have started in 2019, and I'd like to be able to count how many times certain events occurred in say month 4 or month 5 or month 6 of each project, as they roughly last the same amount of time.

I have the Project name and the Received Date of the event. So one idea was to create a column to "lookup" which contains the year and month

Lookup = format('Data'[Received Date],"YYYY-MM")

Using that Lookup, column, I tried to use RankX to rank each month by project.

rankx = rankx(filter('Data','Data'[Project]=earlier('Data'[Project])),'Data'[Lookup],,ASC,Dense)

This worked reasonably well, except, there are some months with no data whatsoever, and they are not included in the rank. This happens a lot at the beginning of the project when there are many months where it is quiet before it ramps up towards the end.

Any thoughts/ideas would be greatly appreciated.

Sample File 

[v-lionel-msft]

Hi @jameschung ,

I agree with @nhoward  answer.

Best regard,
Lionel Chen

[nhoward · Accepted solution]

Hi @jameschung ,

It sounds like you need to know how many months there are between the start date of a project and the data of a transactions/event. 

The DAX function DATEDIFF should be able to help. 

the sample data didn't indicate what your data model is, so i have to make some assumptions.  I assume you have a project table, which has a column for start date.  then you have a transactions table.   You might be able to add a column to the transactions table to show how many months since start date this event happened.  Use your lookup to add a column for the start date, then you can use datediff to get the month number of the project. 

Month of Project = DATADIFF('Data'[Start Date],'Data'[Recieved Date],Month)

[jameschung]

Thanks @nhoward. I hadn't considered that approach and it's useful for another problem I was having too! Much appreciated.