Get certified for free when you join Fabric Data Days 2026 and dive into Fabric, Power BI, SQL, AI, and other essential data skills.
Join nowTry your skills in the Power BI Dataviz World Championship! Round one ends June 26. Join now
There are few columns in data. I want to get the latest non-null value from these columns.
The NewOutputCol is my output column.
In first row, L1 column has the latest non-null value and therefore, output is "A".
In second row, L2 column has the latest non-null value and therefore, output is "E".
In thirs row, L3 column has the latest non-null value and therefore, output is "Z".
In real data, I have 10 such columns to scan from and over 90,000 rows.
Sample data:
| L1 | L2 | L3 | NewOuputCol |
| A | null | null | A |
| D | E | null | E |
| X | null | Z | Z |
Solved! Go to Solution.
Hi @PC20
In Power Query, you can create this custom column
List.Last(List.Select(Record.FieldValues(_), each _ <> null))
Hi @PC20
In Power Query, you can create this custom column
List.Last(List.Select(Record.FieldValues(_), each _ <> null))
@PC20 Probably best if you unpivot your columns and add and index perhaps. Otherwise, maybe this:
Column =
VAR __Data = [L1] & "|" & [L2] & "|" & [L3]
VAR __Table =
ADDCOLUMNS(
GENERATESERIES(1,3,1),
"__Data", PATHITEM(__Data,[Value],TEXT)
)
VAR __Max = MAXX(FILTER(__Table,[__Data] <> BLANK()),[Value])
VAR __Result = MAXX(FILTER(__Table,[Value] = __Max),[__Data])
RETURN
__Result
Don't miss out on Data Days, June 15 through August 7. Learn Fabric, Power BI, SQL, AI and more.
Check out the May 2026 Power BI update to learn about new features.
| User | Count |
|---|---|
| 24 | |
| 22 | |
| 21 | |
| 20 | |
| 14 |
| User | Count |
|---|---|
| 59 | |
| 52 | |
| 41 | |
| 33 | |
| 32 |