Minimum date for family

[Anonymous]

Hello,
I would get a minimum order date on a customer regardless of the family.
Example  :  on the second table, I would only on the two families the minimum date of the client ie the 08 /04/2018 and not two dates.
I used the following formula 

min({<[Big Family]= >} Date)

 but it does not work

Thanks for your help

[OmarBenSalem]

Maybe

min( TOTAL {<[Big Family]= >} Date)

[sunny_talwar]

May be this

Aggr(Min(TOTAL <[ID Customer]> Date), [ID Customer], [Big Family])

[fredericmante]

Bonjour,

Try this:

Date(Min(Aggr(min(Date), [Big Family],[ID customer])))

[Anonymous]

thanks a lot 

your formula works at retail 

if ( Aggr(Min(TOTAL <[ID Customer]> Date), [ID Customer], [Big Family])=min(Date), 1,0)

but not overall, I have the same results between your formula and mine

your formula

sum( {<[ID Customer] = {"=Aggr(Min(TOTAL <[ID Customer]> Date), [ID Customer], [Big Family])=min(Date) "} >} Occurrence_Client)

mine

sum( {<[ID Customer] = {"=min(Date) "} >} Occurrence_Client)

 
it is possible to avoid the "aggr" function for large volumes of data is not recommended

thanks for your help

[sunny_talwar]

I don't think you can use set analysis here... may be you need to use FirstSortedValue() function here. Can you elaborate on your goal? What exactly are you trying to capture? 

[Anonymous]

My goal is to know each client's first order date and attach this order date to the family to make a count. On the other hand, the customer will have to be counted only once. Example if the customer 0001 has a date of order on the 01/08/2018 on the family "A" and another date of order on 11/10/2018 on the family " B ".
This customer must have the value 1 in the family "A" and have the value zero in the family "B".
Is it clearer for you?
Thank you

[Anonymous]

if it can help you attach my work base

thanks

[sunny_talwar · Accepted solution]

Is this what you want?

Expression

Sum(Aggr(If(Min(TOTAL <[ID Customer]> Date) = Date, Occurrence_Client), [ID Customer], [Big Family], Date))

[Anonymous]

Perfect

Thanks a lot for your help