count customers with digital transaction ratio bigger than in a date period

[giorgiokatr]

i have this table

customer	date	transaction type	amount
1	1/1/2022	digital	10
1	1/2/2022	digital	20
1	1/3/2022	digital	30
1	1/4/2022	digital	11
1	1/5/2022	digital	48
1	1/6/2022	digital	69
1	1/1/2022	physical	45
1	1/2/2022	physical	85
1	1/3/2022	physical	96
1	1/4/2022	physical	47
1	1/5/2022	physical	6
1	1/6/2022	physical	22
2	1/1/2022	digital	10
2	1/2/2022	digital	20
2	1/3/2022	digital	30
2	1/4/2022	digital	11
2	1/5/2022	digital	48
2	1/6/2022	digital	69
2	1/1/2022	physical	45
2	1/2/2022	physical	85
2	1/3/2022	physical	96
2	1/4/2022	physical	47
2	1/5/2022	physical	6
2	1/6/2022	physical	22
2	1/1/2022	digital	10
2	1/2/2022	digital	20
2	1/3/2022	digital	30
2	1/4/2022	digital	11
2	1/5/2022	digital	48
2	1/6/2022	digital	69
2	1/1/2022	physical	45
2	1/2/2022	physical	85
2	1/3/2022	physical	96
2	1/4/2022	physical	47
2	1/5/2022	physical	6
2	1/6/2022	physical	22

for each customer i want to calculate the ratio digital / total transactions (digital+physical) amount in the last 3 months

a customers that has ratio >0.5 (digital transactions more than 50% of his total transactions is a digital customer

I want when selecting a month form a slicer (example 01/06/2022) to count the customers of this month that have a >0.5 in the last 3 months of this selection.

[CNENFRNL · Accepted solution]

[giorgiokatr]

can the dax be modified so it shows in a card the distinct count of customers for this? (>0.5)?

[CNENFRNL]

To obtain the same visual effect as Card,

[giorgiokatr]

I would like to count the customers. Example 4 customers in this month had digital ratio (3 month  transactions) >0.5

[CNENFRNL · Accepted solution]

=
COUNTROWS(
    FILTER(
        VALUES( TXN[customer] ),
        CALCULATE(
            CALCULATE( [Total], TXN[transaction type] = "digital" ) / [Total],
            DATESINPERIOD( DATES[Date], MAX( DATES[Date] ), -3, MONTH )
        ) > .5
    )
)

[giorgiokatr]

Thanks worked great!

[CNENFRNL · Accepted solution]