Forum Discussion
Rank without measure.
Hi,
Is it possible to rank a column without a measure value?
This is what I am trying to achieve.
Column 1 Column2 Rank
Product1 US 1
Product1 UK 2
Product1 AU 3
Product2 Uk 1
Product3 AU 1
Rank CC =
RANKX (
SUMMARIZE (
FILTER ( 'Table', 'Table'[Product] = EARLIER ( 'Table'[Product] ) ),
'Table'[Country]
),
CALCULATE ( MAX ( 'Table'[Country] ) ),
,
DESC
)
6 Replies
- Jihwan_KimSuper User
Rank CC =
RANKX (
SUMMARIZE (
FILTER ( 'Table', 'Table'[Product] = EARLIER ( 'Table'[Product] ) ),
'Table'[Country]
),
CALCULATE ( MAX ( 'Table'[Country] ) ),
,
DESC
)- varung8899Helper II
I am also getting all the values of Rank as 1 and I think the formula is incorrect. Also how will the formula be if the product column contains integer values instead of Text ? Instead of Pro01 if it has 1. Pls let me know. TIA.
- smpa01Community Champion
PBIfanatic do you need cal.column / measure?
Ranking within each subset with a RANKX measure indeed requires utilizing another measure inside it to my knowledge.
_rank = //_maxCountry = MAX('Table'[Column2 ]) VAR _0 = MAX('Table'[Column 1 ]) VAR _1 = RANKX(FILTER(ALLSELECTED('Table'),'Table'[Column 1 ]=_0),[_maxCountry],,ASC) RETURN _1 - PBIfanaticHelper V
- Jihwan_KimSuper User
Hi,
I am not sure if you are using it for a calculated measure or a calculated column.
As clearly shown in the picture, I created a calculated column.
You need different formula if you need a calculated measure.
- v-easonf-msftCommunity Support
Hi, PBIfanatic
Regarding the sample data you provided, I did not find any errors in Jihwan_Kim formula.
Please check my attached sample file for more details.
If your problem hasn't been solved, please share more details or related screenshot for further research.
Best Regards,
Community Support Team _ Eason