Countrows(Values confusion

[B00m3r]

As part of my finance model I have a table Purchase Orders with several columns of which one is [PO Number] as below. 

Purchase Orders Table

If I go to PowerBI data area and click on PO Number I see the Table: references with distinct values as shown below.

Distinct Values as shown

If I create a table and drag Purchase Orders into Visual level filters & select Count (Distinct) [see below] i get the result 102

Count (Distinct)

If I create the measure Count rows = COUNTROWS(VALUES('Purchase Orders'[PO NUMBER] )) i get the number 103

The DATA comes from an excel spreadsheet.

I'm trying to udnerstand what countrows is doing and whether this is the incorrect measure to use to count rows?

Regards B00M3R

[MattAllington]

 countrows is the correct formula, but you shouldn't need values. Just count the rows of the table (given po is unique).

This is very strange. Any chance that you have two tables that contain PO number? 

[B00m3r]

Hi Matt

Yes i have discovered it in the main data fact table. Would this be the issue?

[v-yulgu-msft]

Hi @B00m3r,

In my test, COUNTROWS function returns correct result, the same as selecting Count(distinct) for field. So, I think this problem is related to specific data. Could you provide more sample data or share .pbix file directly?

Besides, please try below measure to check whether it returns 102 or 103.

Count rows = DISTINCTCOUNT('Purchase Orders'[PO NUMBER] )

Best regards,
Yuliana Gu