Inconsistent Measure using Switch

[SamuelROS]

Hello Experts,

I’m trying to replicate the desired output from a working Calculated column into a measure using switch. Although my per line items are getting the desired result, the totals are a little inconsistent.

Calculated Column

CC Age = 
SWITCH(
    TRUE(),
    DIM_MRF[Quantity] = DIM_MRF[QuantityDelivered],
    DATEDIFF(
        RELATED( DIM_PO[Date].[Date] ),
        DIM_MRF[DeliveryDate].[Date],
        DAY
    ),
    DIM_MRF[Quantity] <> DIM_MRF[QuantityDelivered],
    DATEDIFF(
        RELATED( DIM_PO[Date].[Date] ),
        TODAY(),
        DAY
    )
)

Measure

M Age = 
MAXX('DIM_MRF',
    SWITCH(
        TRUE(),
        SELECTEDVALUE(DIM_MRF[Quantity]) = DIM_MRF[QuantityDelivered],
            DATEDIFF(
                RELATED(DIM_PO[Date].[Date]),
                DIM_MRF[DeliveryDate].[Date],DAY
            ),
        SELECTEDVALUE(DIM_MRF[Quantity]) <> DIM_MRF[QuantityDelivered],
            DATEDIFF(
                RELATED(DIM_PO[Date].[Date]),
                TODAY(),DAY
            )
    )
)

MRFNumber	PONumber	MaterialNumber	POItemNumber	Quantity	QuantityDelivered	POUnitPrice	DeliveryDate
IST-22-52	PO-00001	IST-00131	1	5	5	422	17/11/2022
IST-22-52	PO-00001	IST-00135	5	5		12
IST-22-52	PO-00001	IST-00136	6	2		3380
IST-22-52	PO-00001	IST-00137	7	2		608
IST-22-52	PO-00001	IST-00138	8	2		172
IST-22-52	PO-00001	IST-00139	9	2		708
IST-22-52	PO-00001	IST-00140	10	2		134
IST-22-52	PO-00001	IST-00141	11	5	5	1146	17/11/2022
IST-22-46	PO-00002	IST-00104	1	20	20	82	15/11/2022
IST-22-46	PO-00002	IST-00102	3	10	10	556	15/11/2022
IST-22-46	PO-00002	IST-00101	4	5	5	50	15/11/2022
IST-22-46	PO-00002	IST-00100	5	1	1	740	15/11/2022
IST-22-46	PO-00002	IST-00099	6	15	11	61.86	15/11/2022
IST-22-46	PO-00002	IST-00098	7	2	2	1260	25/11/2022
IST-22-46	PO-00002	IST-00094	11	5	5	9	15/11/2022
IST-22-46	PO-00002	IST-00091	14	1		95.22
IST-22-41	PO-00003	IST-00158	1	5	5	426.5	16/11/2023
IST-22-41	PO-00003	IST-00159	2	4	4	358.26	16/11/2023
IST-22-41	PO-00003	IST-00160	3	1	1	307.08	16/11/2023
IST-22-41	PO-00003	IST-00161	4	4	4	196.19	16/11/2023
IST-23-81	PO-00004	IST-2023-468	1	6		5385.61
IST-23-81	PO-00004	IST-2023-467	2	6		1520
IST-23-55	PO-00005	IST-2023-547	2	3		1265
IST-23-55	PO-00005	IST-2023-548	3	6		255
IST-23-55	PO-00005	IST-2023-549	4	6		255
IST-23-55	PO-00005	IST-2023-550	5	3		198
IST-23-55	PO-00005	IST-2023-553	8	2		1848.75
IST-23-55	PO-00005	IST-2023-555	10	1		2985
IST-23-55	PO-00005	IST-2023-546	1	3		5575
IST-23-74	PO-00006	IST-00269	2	2		13.48
IST-23-74	PO-00006	IST-00268	3	40		12.52
IST-23-74	PO-00006	IST-00267	1	2		30

Description	PONumber	Date	Currency	Vendor
Description 01	PO-00001	16/08/2022	USD	Vendor 1
Description 02	PO-00002	27/09/2022	EUR	Vendor 2
Description 03	PO-00003	09/05/2023	EUR	Vendor 2
Description 04	PO-00004	05/06/2023	EUR	Vendor 2
Description 05	PO-00005	04/07/2023	EUR	Vendor 2
Description 06	PO-00006	27/09/2022	USD	Vendor 1

[Chris_White · Accepted solution]

Remove the SELECTEDVALUE function from your measure.  MAXX is already converting to row context so having SELECTEDVALUE in there is tripping it up.

M Age =

MAXX('DIM_MRF',

    SWITCH(

        TRUE(),

        DIM_MRF[Quantity] = DIM_MRF[QuantityDelivered],

            DATEDIFF(

                RELATED(DIM_PO[Date].[Date]),

                DIM_MRF[DeliveryDate].[Date],DAY

            ),

        DIM_MRF[Quantity] <> DIM_MRF[QuantityDelivered],

            DATEDIFF(

                RELATED(DIM_PO[Date].[Date]),

                TODAY(),DAY

            )

    )

)

[SamuelROS]

I feel stupid. But thank you so much!