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
Hi all,
I have a bank dimension table like this
Cod_bank | Desc_bank |
30 | BANK_A |
61 | BANK_B |
182 | BANK_C |
198 | BANK_D |
My question is regarding fact table (with snapshots)
Cod_bank | date_update | Registers |
30 | 02/01/2011 | 107.449 |
30 | 19/01/2011 | 123.572 |
30 | 23/01/2011 | 123.703 |
30 | 30/01/2011 | 123.958 |
30 | 06/02/2011 | 124.670 |
61 | 15/06/2011 | 2.837 |
182 | 23/01/2011 | 436.329 |
182 | 30/01/2011 | 435.971 |
182 | 06/02/2011 | 441.024 |
198 | 16/01/2011 | 45 |
198 | 23/01/2011 | 53 |
198 | 30/01/2011 | 74 |
198 | 06/02/2011 | 85 |
198 | 20/02/2011 | 80 |
198 | 13/03/2011 | 79 |
198 | 27/03/2011 | 77 |
My requirement is to have a date_update filter and get the number of Registers of the max date_update by Bank.
DAX Measures
max_date = max(Facts[date_update])
num_registers = SUM(Facts[Registers])
Registers on max date for this bank = CALCULATE([num_registers];
FILTER(Facts;Facts[date_update]=[max_date]))
Filter is failing due to context, I need max_date for Each bank before date selected
Report configuration with date_update, Before filter .
Desired output is the number of Registers in the max available date Before selected, for Each bank
If Selected 15/06/2011 (June 15 2011)
Des_bank | date_update | Registers |
BANK_A | 06/02/2011 | 124.670 |
BANK_B | 15/06/2011 | 2.837 |
BANK_C | 06/02/2011 | 441.024 |
BANK_D | 27/03/2011 | 77 |
Current report for 15/06/2011 (June 15 2011)
If Selected 20/01/2011 (January 20 2011)
Des_bank | date_update | Registers |
BANK_A | 06/02/2011 | 123.572 |
BANK_D | 27/03/2011 | 45 |
Current Report for Selected 20/01/2011 (January 20 2011)
Kind Regards
Solved! Go to Solution.
Hi @dpombal,
What about the formula below?
Measure 2 = SUMX('DIM_BANKS',[Measure])
Here is the result output.
Best Regards,
Cherry
Hi @dpombal,
Do you want to get the ouput below?
If this is your desired output, please refer to the formula below.
Measure =
CALCULATE (
SUM ( fact_[Registers] ),
FILTER ( 'fact_', 'fact_'[date_update] = MAX ( 'fact_'[date_update] ) )
)
If you still need help, please share your desired output so that we could help further on it.
Best Regards,
Cherry
Hi ,
your solutions works OK for me except on grand totals
See attached report
https://drive.google.com/file/d/10BliwwN3ZAd8PibCGkz7k07fZDPUn6fK/view?usp=sharing
Something like
IF HASONEVALUE BANK
THEN CURRENT behaviour of Measure
ELSE in totals SUM PREVIOUS ROWS with registers on max date available for each bank
Regards
Hi @dpombal,
What about the formula below?
Measure 2 = SUMX('DIM_BANKS',[Measure])
Here is the result output.
Best Regards,
Cherry
Hi, I am surprised the approach using
below 2 measures works fine,
Measure_aux = CALCULATE ( [sum_registers],
FILTER ( Facts, Facts[date_update] = MAX ( Facts[date_update] ) )
)
Measure 2 = SUMX('DIM_BANKS',[Measure_aux])
However, after mixing both in a single measure...something strange happens
Measure3 = SUMX('DIM_BANKS',
CALCULATE (
[sum_registers],
FILTER ( Facts, Facts[date_update] = MAX ( Facts[date_update] ) )
)
)
The sum is not correct....I can't figure out why this is happening and the order of evaluation... but this is the strange result
Regards
Hi @dpombal,
I'm afraid that you cannot mix the formulas like that. If you nest a measure in measure, it may cause incorrect result.
For the reason, you could have a reference of this blog.
Best Regards,
Cherry
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 | |
| 23 | |
| 21 | |
| 17 | |
| 13 |
| User | Count |
|---|---|
| 58 | |
| 50 | |
| 38 | |
| 29 | |
| 24 |