Sales dollars between two dates

[BORUCH]

hi all
i have my sales history table with approx. 10 mil rows of invoice and credit memo history
it has columns like customer number, customer name ,invoice number, invoice date ,item number, item description ,qty ,amount ... etc..

what I'm trying to do is have a formula that gives me sales between two dates that i want to select in a slicer so i created two disconnected date tables one of them is called P1 Compare From Date the other is called P1 Compare To Date then i have a formula in my sale history table as follows Total Sales$ Sold From -- to P1:=SUMX(FILTER('salesdata', 'salesdata'[Invoice Date] >= MIN('P1 Compare From Date'[Date]) && 'salesdata'[Invoice Date] <= MAX('P1 Compare To Date'[Date])), 'salesdata'[AMOUNT])

the problem is it runs extremely slow and i have a very quick computer any help or recommendations would be appreciated

 

[Matt Allington]

You are on the right track but you need some changes. One issue is that you are using a Virtual filter. There are no relationships between your tables meaning the engine basically has to trawl through millions of rows to find the data. You have to do it, but there are better ways. Do the following

1. Get a calendar table

Build a Reusable Calendar Table with Power Query

In this step by step guide you'll learn how to create a Calendar Table using Power Query and save the code to reuse it later.

exceleratorbi.com.au

2. Use it as shown in this article The Optimal Shape for Power BI Data joining it to the date column in your transaction table.

3. Then you put your filter on the calendar table, not the transaction table

=
CALCULATE (
    SUM ( 'salesdata'[AMOUNT] ),
    FILTER (
        'Calendar',
        'Calendar'[Date] >= MIN ( 'P1 Compare From Date'[Date] )
            && 'Calendar'[Date] <= MAX ( 'P1 Compare To Date'[Date] )
    )
)

Using calculate ensures that only the data you need remains before the calculation starts; this is best practice

 

[BORUCH]

You are on the right track but you need some changes. One issue is that you are using a Virtual filter. There are no relationships between your tables meaning the engine basically has to trawl through millions of rows to find the data. You have to do it, but there are better ways. Do the following

1. Get a calendar table

Build a Reusable Calendar Table with Power Query

In this step by step guide you'll learn how to create a Calendar Table using Power Query and save the code to reuse it later.

exceleratorbi.com.au

2. Use it as shown in this article The Optimal Shape for Power BI Data joining it to the date column in your transaction table.

3. Then you put your filter on the calendar table, not the transaction table

=
CALCULATE (
    SUM ( 'salesdata'[AMOUNT] ),
    FILTER (
        'Calendar',
        'Calendar'[Date] >= MIN ( 'P1 Compare From Date'[Date] )
            && 'Calendar'[Date] <= MAX ( 'P1 Compare To Date'[Date] )
    )
)

Using calculate ensures that only the data you need remains before the calculation starts; this is best practice

thanks that worked

 

[smozgur]

thanks that worked

I marked the solution post accordingly. That would be great if you could do the same in your future questions to help future readers.