Rank within double grouping - possible?

[reinholz]

Hey,

I'm trying to add a rank column to a fact table that ranks based on two groupings.

My table basically contains records for user activities with user id, created date and a column with the product id for which the activity happend.

It looks like this:

  

What I now want to do is to create a column that ranks the activities by created date, based on user AND product.

How can I do this?

Thanks!

[OwenAuger · Accepted solution]

Hi @reinholz

If I've understood you correctly, you want a calculated column that gives you the current row's [created at] rank, among all rows with the same [user id] and [product id].

With DAX, you can do that with an expression like this (you may want to change ASC to DESC):

Rank =
RANKX (
    CALCULATETABLE (
        FactTable,
        ALLEXCEPT ( FactTable, FactTable[user id], FactTable[product id] )
    ),
    FactTable[created at],
    ,
    ASC
)

[Anonymous]

Hi @aowen,

The solution works great  for static content.

However,  I cannot use Calculated Columns as there are 4 other fileds and a calendar field that are used for filtering data (dynamic filters) on the report.

Also, I need exactly 10 entries per group.

For example, in the above scenario

for PRODUCT 1, there shall be 10 customers ordered by Sales Amount,

for PRODUCT 2, there shall be 10 Customers ..etc

I tried with the below DAX expressions along with another field like country ..etc in a matrix and applied.

But as in snapshot, I m seeing incorrect values

The rank values are coming up from 2 instead of 1

What am I doing wrong?

//Rank decimal values

Rank over Rank of Customers = IF (HASONEVALUE( FACT_Mail[Sender]),
RANKX(ALL(FACT_Mail[Sender]), [Rank of Customers by Request Received], ,DESC,DENSE) , BLANK())

//Measure to generate decimal value by taking into account Primary and Secondary Measure to avoid ties

Rank of Customers by Request Received =

IF (HASONEVALUE(FACT_Mail[Sender]),

RANKX(ALL(FACT_Mail[Sender]), [Total Received Requests], ,DESC,DENSE)
+ DIVIDE (
RANKX ( ALL ( FACT_Mail ), [Total Mail Size]+0,, DESC, Dense ),
( COUNTROWS ( ALL ( FACT_Mail ) ) + 1 )),
BLANK())

//Primary Measure

Total Received Requests = CALCULATE(COUNT(FACT_Mail[RowId]), USERELATIONSHIP(DIM_DateTable[DateKey], FACT_Mail[DateKey_DateTimeReceived]) )

//Secondary Measure

Total Mail Size = DIVIDE(CALCULATE(SUM(FACT_Mail[Size]), USERELATIONSHIP(DIM_DateTable[DateKey], FACT_Mail[DateKey_DateTimeReceived]) ), 1024, 0)

Thanks,

Mannu

[reinholz]

Thank you! Works exactly as expected!