Don't miss your chance to take the Fabric Data Engineer (DP-700) exam on us!
Learn moreThe FabCon + SQLCon recap series starts April 14th at 8am Pacific. If you’re tracking where AI is going inside Fabric, this first session is a can't miss. Register now
I have two tables:
Products table has a single column with one product per field:
List table has several columns but the column I'm interested in has fields that can have as little as one product to as much as 17, all separated with a comma and no spaces in between:
The issue I'm facing and need help with is to correctly count each product either, ideally as as a measure. I tried two measures but they both gave me an incorrect count of blanks; there are no blank values either table.
What I found is that both measures count only the *first* value they find in the List column and then continue counting subsequent rows:
How can I correctly count all values in each field from the List table and have zero "blank" counts?
Thanks!
Solved! Go to Solution.
@Anonymous
Please delete or deactivate that relationship or use
Customer Products at M0 =
CALCULATE (
SUMX (
VALUES ( Products[Product] ),
SUMX (
VALUES ( Query1[M0_LIST] ),
INT ( CONTAINSSTRING ( Query1[M0_LIST], Products[Product] ) )
)
),
CROSSFILTER ( Query1[M0_LIST], Products[Product], NONE )
)
Hi @Anonymous
If the same product is repeated multiple times inside the string and you want to consider that in the counting then please use
Customer Products at M0 =
SUMX (
VALUES ( Products[Product] ),
SUMX (
VALUES ( Query1[M0_LIST] ),
VAR String = Query1[M0_LIST]
VAR Items =
SUBSTITUTE ( String, ",", "|" )
VAR Length =
COALESCE ( PATHLENGTH ( Items ), 1 )
VAR T1 =
GENERATESERIES ( 1, Length, 1 )
VAR T2 =
SELECTCOLUMNS ( T1, "@Item", PATHITEM ( Items, [Value] ) )
RETURN
COUNTROWS ( FILTER ( T2, [@Item] = Products[Product] ) )
)
)
Hi @tamerj1 , thank you. I tried using the first measure and still gave me the incorrect counts, using the second one gave me this result.
The same product is not repeated inside the string, but two clients can definitely have the same products but they would be listed in different rows.
@Anonymous
Can you please explain exactly what is the problem?
Hi @tamerj1 , with your second measure I got a count of 1 for each product:
Also, the List table contains blank values/rows too, and as for active relationships I see that the Product and Lists table are connected in the Product[Products] column and the List[M0_LIST] column, one to many relationship. I don't know if this is causing the issue and hope this gives more clarity to my issue.
@Anonymous
Please delete or deactivate that relationship or use
Customer Products at M0 =
CALCULATE (
SUMX (
VALUES ( Products[Product] ),
SUMX (
VALUES ( Query1[M0_LIST] ),
INT ( CONTAINSSTRING ( Query1[M0_LIST], Products[Product] ) )
)
),
CROSSFILTER ( Query1[M0_LIST], Products[Product], NONE )
)
Hi @Anonymous
please try
Customer Products at M0 =
SUMX (
VALUES ( Products[Product] ),
SUMX (
VALUES ( Query1[M0_LIST] ),
INT ( CONTAINSSTRING ( Query1[M0_LIST], Products[Product] ) )
)
)
If you have recently started exploring Fabric, we'd love to hear how it's going. Your feedback can help with product improvements.
A new Power BI DataViz World Championship is coming this June! Don't miss out on submitting your entry.
Share feedback directly with Fabric product managers, participate in targeted research studies and influence the Fabric roadmap.
| User | Count |
|---|---|
| 9 | |
| 6 | |
| 3 | |
| 2 | |
| 2 |
| User | Count |
|---|---|
| 21 | |
| 14 | |
| 11 | |
| 6 | |
| 5 |