Forum Discussion

paweldm's avatar
paweldm
Helper II
9 years ago
Solved

segmentation on days difference between two dates

Hi,

 

Problem: how can I perform segmentation of customers based on the DiffDates (number of days between today and date of the last transaction) with use of ranges defined in a separate and non-conected with the model table. I would like to select for example a "Hot"  range on the slicer to filter all my customers who made last transaction within 0-30 days from today.

 

I have the following input:

 

Ranges Table (it is not conected with the model)

Ranges[Name] | Ranges[Min Value in Days] | Ranges[Max Value in Days]

Hot | 0 | 30
Warm| 30 | 60

Cold |60 | 180

Frozen |180 | 9999999

 

Sales Table
Sales[CustomerId] | Sales[InvoiceNo] | Sales[Date]

 

Mesures

#LastTransaction = LASTDATE(Sales[Date])

#DiffDates = if(ISBLANK([#LastTransaction]);BLANK();VALUE(TODAY()-[#LastTransaction]))

 

I have the following tabel report:

 

Sales[CustomerId] | #LastTransaction | #DiffDates

 

I need to filter for example only HOT range customers by selecting a range with use of filter/slicer.

 

Thx a lot for help !

 

Paweł

 

 

 

  • Hi paweldm,

     

    In your scenario, you can create a new table from the Sales table:

     

    Table = SUMMARIZE('Table1','Table1'[CustomerID],"LastTransaction",MAX('Table1'[date]))

     

    Then create calculated columns in this table:

    DiffDates = if(ISBLANK([LastTransaction]),BLANK(),VALUE(TODAY()-[LastTransaction]))

     

    Range = SWITCH(TRUE(),'Table'[DiffDates]>0 && 'Table'[DiffDates]<30,"Hot",'Table'[DiffDates]>30 && 'Table'[DiffDates]<60,"Warm",'Table'[DiffDates]>60 && 'Table'[DiffDates]<180,"Cold",'Table'[DiffDates]>180 && 'Table'[DiffDates]<9999999,"Frozen")

     

    Create a relationship between the Sales table and this new table use CustomerID. Drag Range column from the new table into a slicer visual. For more information, please check attached .pbix.

     

     

    Best Regards,
    Qiuyun Yu

3 Replies

  • v-qiuyu-msft's avatar
    v-qiuyu-msft
    Community Support

    Hi paweldm,

     

    In your scenario, you can create a new table from the Sales table:

     

    Table = SUMMARIZE('Table1','Table1'[CustomerID],"LastTransaction",MAX('Table1'[date]))

     

    Then create calculated columns in this table:

    DiffDates = if(ISBLANK([LastTransaction]),BLANK(),VALUE(TODAY()-[LastTransaction]))

     

    Range = SWITCH(TRUE(),'Table'[DiffDates]>0 && 'Table'[DiffDates]<30,"Hot",'Table'[DiffDates]>30 && 'Table'[DiffDates]<60,"Warm",'Table'[DiffDates]>60 && 'Table'[DiffDates]<180,"Cold",'Table'[DiffDates]>180 && 'Table'[DiffDates]<9999999,"Frozen")

     

    Create a relationship between the Sales table and this new table use CustomerID. Drag Range column from the new table into a slicer visual. For more information, please check attached .pbix.

     

     

    Best Regards,
    Qiuyun Yu

    • paweldm's avatar
      paweldm
      Helper II
      Thank you ! It is brilliante solution.
    • paweldm's avatar
      paweldm
      Helper II

      v-qiuyu-msft wrote:

      Hi paweldm,

       

      In your scenario, you can create a new table from the Sales table:

       

      Table = SUMMARIZE('Table1','Table1'[CustomerID],"LastTransaction",MAX('Table1'[date]))

       

      Then create calculated columns in this table:

      DiffDates = if(ISBLANK([LastTransaction]),BLANK(),VALUE(TODAY()-[LastTransaction]))

       

      Range = SWITCH(TRUE(),'Table'[DiffDates]>0 && 'Table'[DiffDates]<30,"Hot",'Table'[DiffDates]>30 && 'Table'[DiffDates]<60,"Warm",'Table'[DiffDates]>60 && 'Table'[DiffDates]<180,"Cold",'Table'[DiffDates]>180 && 'Table'[DiffDates]<9999999,"Frozen")

       

      Create a relationship between the Sales table and this new table use CustomerID. Drag Range column from the new table into a slicer visual. For more information, please check attached .pbix.

       

       

      Best Regards,
      Qiuyun Yu


      Is there a solution to add to the report one more measure to calculate frequency of invoices?

       

      The frequency is to be calculated as a Customer LifeTime (calculated in days since the first transaction till today)  divided by TotalInvoices (calculated as a total number of invoices generated by a customer during its LifeTime)

       

      My current tabel report has the following columns and works exelent with "temperature" slicer:

      CustomerID | LastTransaction | DiffDates