Power BI is turning 10! Tune in for a special live episode on July 24 with behind-the-scenes stories, product evolution highlights, and a sneak peek at what’s in store for the future.
Save the dateEnhance your career with this limited time 50% discount on Fabric and Power BI exams. Ends August 31st. Request your voucher.
Hi All,
This might be a very basic question to many of you. I am trying to create a model where all my transactions carried out in different currencies is translated into USD using montly exchange rates.
My model has three tables, 1) calendar (marked as date table), 2) fact_transaction table (this table has transaction values as well as transation date) and 3) exchange rate table (containing monthly average exchange rates on 1st date of each month against multiple currencies).
Relationship is direct between calendar and transaction date as well as calendar and monthly exchange rate date.
Help to bring clarity would be appreciated.
Solved! Go to Solution.
You could do it as a measure like
USD Amount Measure =
SUMX (
'Transactions',
VAR CurrencyRate =
LOOKUPVALUE (
'Exch rate'[rate],
'Exch Rate'[Date],
DATE ( YEAR ( 'Transactions'[Transaction date] ), MONTH ( 'Transactions'[Transaction date] ), 1 ),
'Exch rate'[Currency], 'Transactions'[Currency]
)
RETURN
'Transactions'[Amount] * CurrencyRate
)
but I think performance would likely be better having a calculated column and then building measures based on that. If it was a column then it would be calculated once, during data refresh, and then you could have measures to perform sum, average etc on the calculated column without having to pay the cost of all the lookups.
Best advice is to try both ways and use DAX Studio to compare performance
You could create a calculated column in the transactions table like
USD Amount =
VAR CurrencyRate =
LOOKUPVALUE (
'Exch rate'[rate],
'Exch Rate'[Date],
DATE ( YEAR ( 'Transactions'[Transaction date] ), MONTH ( 'Transactions'[Transaction date] ), 1 ),
'Exch rate'[Currency], 'Transactions'[Currency]
)
RETURN
'Transactions'[Amount] * CurrencyRate
Hi John,
This did work. But i was wondering if instead of adding conditional column, there was a way to achieve this through dax measures.
Since i intend to add more factual transaction table and as far as i know measures are way better in terms of performance than calculated columns.
You could do it as a measure like
USD Amount Measure =
SUMX (
'Transactions',
VAR CurrencyRate =
LOOKUPVALUE (
'Exch rate'[rate],
'Exch Rate'[Date],
DATE ( YEAR ( 'Transactions'[Transaction date] ), MONTH ( 'Transactions'[Transaction date] ), 1 ),
'Exch rate'[Currency], 'Transactions'[Currency]
)
RETURN
'Transactions'[Amount] * CurrencyRate
)
but I think performance would likely be better having a calculated column and then building measures based on that. If it was a column then it would be calculated once, during data refresh, and then you could have measures to perform sum, average etc on the calculated column without having to pay the cost of all the lookups.
Best advice is to try both ways and use DAX Studio to compare performance
User | Count |
---|---|
25 | |
12 | |
8 | |
6 | |
6 |
User | Count |
---|---|
26 | |
12 | |
12 | |
10 | |
6 |