How to do it...

To implement this recipe, do the following:

1. Create the following measures:

Lost Customers = 
    VAR __Threshold = 1
    VAR __ThresholdUnit = "YEARS"
    VAR __Month = MAX('R03_Dates'[Month])
    VAR __Year = MAX('R03_Dates'[Year])
    VAR __MonthStart = DATE(__Year, __Month, 1)
    VAR __MonthEnd = MAX('R03_Dates'[Date])
    VAR __PreviousCustomers = 
        SUMMARIZE(
            FILTER(
                'R03_Table',
                [Date] < __MonthEnd
            ),
            [Customer],
            "__LastPurchase",MAX([Date])
        )
    VAR __PreviousCustomers1 = 
        ADDCOLUMNS(
            __PreviousCustomers,
            "__LostDate",
            SWITCH(
                __ThresholdUnit,
                "DAYS",[__LastPurchase] + __Threshold,
                "MONTHS",EOMONTH([__LastPurchase],__Threshold),
                "YEARS",DATE(YEAR([__LastPurchase]) + __Threshold,MONTH([__LastPurchase]),1)
            )
        )
    VAR __LostCustomers = 
        DISTINCT(
            SELECTCOLUMNS(
                FILTER(
                    __PreviousCustomers1,
                    MONTH([__LostDate]) = __Month && 
                        YEAR([__LostDate]) = __Year),
                "__Customer",[Customer]
            )
        )
RETURN
    COUNTROWS(__LostCustomers)

2. Create the Recovered Customers measure using the code in the R03_RecoveredCustomers.txt text file located in the GitHub repository.
3. On a Report page, create a matrix visualization and place the Month Name column from R03_Dates into the Rows area, the Year column from R03_Dates into the Columns area, and the Lost Customers measure into the Values area.
4. On a report page, create a matrix visualization and place the Month Name column from R03_Dates into the Rows area, the Year column from R03_Dates into the Columns area, and the Recovered Customers measure into the Values area.