Variable as a single measure

I'm not sure if it's necessarily possible to optimize significantly. Because the context is different each time, you need to evaluate each of the prior months separately. I don't see a way to get around that since you can't actually reuse a computation.

That said, you might be able to get some degree of parallelization if you write it differently. See if this works any better:

Weighted Average =
VAR PriorMonths =
    ADDCOLUMNS (
        GENERATESERIES ( 1, 5 ),
        "MonthDailySales",
            DIVIDE (
                CALCULATE ( [Sold Prod Qnty], DATEADD ( 'Date'[Date], -[Value], MONTH ) ),
                CALCULATE ( [Unique Sales Day Count], DATEADD ( 'Date'[Date], -[Value], MONTH )
                )
            ),
        "Weight",
            ( 7 - [Value] ) / 2
    )
VAR WtdAvg =
    DIVIDE (
        SUMX ( PriorMonths, [Weight] * [MonthDailySales] ),
        SUMX ( PriorMonths, [Weight] )
    )
RETURN
    DIVIDE ( WtdAvg, [Unique Sales Day Count] )