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
How to convert the below sql statement into DAX to calculate Average Lag? Thanks
SELECT avg(isnull(datediff(day, b.ClaimSubmitDate, t.PostDate),0)) as AvgLag
FROM [dbo].[Bill] b join [dbo].[Transactions] t
on b.BillID= t.BillID
WHERE t.[TransactionTypeID] = 2
Hi @JulietZhu
There are two approches.
1.
create a calculated column
related = LOOKUPVALUE(b[ClaimSubmitDate],b[BillID],t[BillI])
Then create a measure
Measure =
VAR Table1 =
NATURALINNERJOIN ( b, t )
VAR datediff =
IF (
ISBLANK ( DATEDIFF ( MAX ( [related] ), MAX ( [PostDate] ), DAY ) ),
0,
DATEDIFF ( MAX ( [related] ), MAX ( [PostDate] ), DAY )
)
RETURN
CALCULATE (
AVERAGE ( 'Table'[datediff] ),
FILTER ( ALL ( 'Table' ), [TransactionTypeID] = 2 )
)
2. create a new table
Table = NATURALINNERJOIN(b,t)
Then in the new table
create calculated columns
datediff =
IF (
ISBLANK ( DATEDIFF ( [ClaimSubmitDate], [PostDate], DAY ) ),
0,
DATEDIFF ( [ClaimSubmitDate], [PostDate], DAY )
)
average = CALCULATE(AVERAGE('Table'[datediff]),FILTER(ALL('Table'),[TransactionTypeID]=2))
Best Regards
maggie
Honestly, any question that starts with "I want to convert this SQL into DAX" generally never go well when all you present is the SQL. Sample data and expected result will get you there quicker. Please see this post regarding How to Get Your Question Answered Quickly: https://community.powerbi.com/t5/Community-Blog/How-to-Get-Your-Question-Answered-Quickly/ba-p/38490
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 |
|---|---|
| 22 | |
| 22 | |
| 21 | |
| 20 | |
| 12 |
| User | Count |
|---|---|
| 58 | |
| 55 | |
| 42 | |
| 36 | |
| 35 |