Reform table

[Anonymous]

I have a data that records the member upgrade assessment of our member as below:

customer   date              prev_member_level    current_member_level

John          1Jan2017       Classic                         Gold

John          4Apr2017      Gold                            Gold 

John         15 Jun2017    Gold                            Platinum

John         1Dec2017      Platinum                      Gold

Mary         2Feb2017      Silver                           Gold

How can i reform the information above to be like table below?

customer   start_date          end_date             member_level

John           1Jan2016          31Dec2016          Classic

John           1Jan2017          14Jun2017           Gold

John           15Jun2017        Today()                Platinum

Mary          1Jan2016           1Feb2017           Silver

Mary          2Feb2017          6Apr2018            Gold

[Phil_Seamark · Accepted solution]

HI @Anonymous

This calculated table is getting close.  There is a bit going on, but I think it addresses most of the quirks.

I have attached a PBIX file as well

New Table = 
VAR RetVal = 
        ADDCOLUMNS(
            'Table',
            "Next Date",
            MINX(
                FILTER(
                    'Table',
                    'Table'[Customer] = EARLIER('Table'[Customer]) && 
                    'Table'[Date] > EARLIER('Table'[Date]) && 
                    [pre_member_level]<> [current_member_level]
                    ),
                    [Date]-1)
                    )
VAR cleanTable = FILTER(RetVal,[current_member_level]<>[pre_member_level])

VAR Step2 =  
    SELECTCOLUMNS(
        cleanTable ,
        "Customer" , [Customer] ,
        "Start_date" , [Date] ,
        "end_date" , if( 
                            ISBLANK([Next Date]),
                            TODAY(),
                            [Next Date]) ,
        "member_level" , [current_member_level]

    )
    
    
VAR FirstLines =
    SELECTCOLUMNS( 
        GENERATE(SELECTCOLUMNS(SUMMARIZECOLUMNS('Table'[Customer],"end_date" , MIN('Table'[Date])),"C",[Customer],"D",[end_date]),filter('Table','Table'[Customer]=[C] && 'Table'[Date]=[D])),
        "Customer",[Customer] ,
        "Start_date",DATE(2016,1,1) ,
        "End_date",[Date]-1,
        "member_level",[pre_member_level]
    )
        
RETURN    UNION( FirstLines,Step2)
    

[Anonymous]

Thank you so much @Phil_Seamark, I got it!