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 nowData Days is here! Join us now for 60+ days of learning, challenges, and connection. Learn more
¿What would be the simplest DAX formula in PowerBI to calculate row differences?
I tried using EARLIER() but it cannot handle the operation at the limit (missing data):
EARLIER/EARLIEST refers to an earlier row context which doesn't exist.
| row | value | diff(value)? |
| 1 | 5 | |
| 2 | 1 | -4 |
| 3 | 6 | 5 |
| 4 | 7 | 1 |
| 5 | 3 | -4 |
| 6 | 3 | 0 |
| 7 | 1 | -2 |
| 8 | 1 | 0 |
| 9 | 8 | 7 |
Hi, @BjoernSchaefer
try this
Difference =
CALCULATE(
SUM('Data'[value]),
FILTER(
ALL('Data'),
'Data'[row] = EARLIER('Data'[row]) - 1
)
) - 'Data'[value]
the code will only work if your row column has unique value. If not then create an index column starting from 0.
If you have missing values in your value column then try this: make sure you have an index column starting from0
Difference =
LOOKUPVALUE (
Data[value],
Data[Index],
Data[Index] - 1,
BLANK ()
) - Data[value]
this thread might help:
Solved: I need output like this by using measure ( calcula... - Microsoft Fabric Community
Proud to be a Super User!
@frankly Or you can use IF statememnt:
Difference =
IF(
Tabelle[row] = 1,
BLANK(),
(LOOKUPVALUE(Tabelle[value], Tabelle[row], Tabelle[row] - 1) - Tabelle[value]) * -1
)
Proud to be a Super User!
Hi @frankly ,
You can try this measure:
diff =
VAR _current_row = SELECTEDVALUE ( Table_[row] )
VAR _current_value = SELECTEDVALUE ( Table_[value] )
VAR _prev_row = MAXX ( FILTER ( ALL ( Table_[row] ), Table_[row] < _current_row ), Table_[row] )
VAR _prev_value = MAXX ( FILTER ( ALL ( Table_ ), Table_[row] = _prev_row ), Table_[value] )
RETURN
IF ( NOT ISBLANK ( _prev_value ), _current_value - _prev_value )
+
For the calculated coumn:
Column =
VAR _current_row = Table_[row]
VAR _current_value = Table_[value]
VAR _prev_row = MAXX ( FILTER ( VALUES ( Table_[row] ), Table_[row] < _current_row ), Table_[row] )
VAR _prev_value = MAXX ( FILTER ( Table_, Table_[row] = _prev_row ), Table_[value] )
RETURN
IF ( NOT ISBLANK ( _prev_value ), _current_value - _prev_value )
If this post helps, then please considerAccept it as the solution to help the other members find it more quickly.
Appreciate your Kudos ![]()
Stand with Ukraine!
Hey frankly,
i was able to do it with SWITCH and LOOKUPVALUE.
Maybe there's a more elegant way.
formula = SWITCH(TRUE(),Tabelle[row]=1,BLANK(),(LOOKUPVALUE(Tabelle[value],Tabelle[row],Tabelle[row]-1)-Tabelle[value])*-1)
Kind Regards
Björn
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 |
|---|---|
| 23 | |
| 21 | |
| 20 | |
| 18 | |
| 13 |
| User | Count |
|---|---|
| 58 | |
| 51 | |
| 37 | |
| 30 | |
| 26 |