Forum Discussion

Anonymous's avatar
Anonymous
Not applicable
4 years ago
Solved

Calculate Phase/Stage duration

Hi Dear Power Bi Helpers,

 

I need some advice. I would like to calculate the Phase duration for all candidates in my data .

Is there any proper way how to solve it?

 

This is how my data looks (please see my photo attached).

 

Explanation:

Inbox - Reviewing (11 May-14 May))-4 days

Reviewing -1st Interview (14 May) - 1 day

1st Interview -Selection1 (14 May-19 May) - 6 days

 

...and so on

till the final hiring stage 14 june for this person.

 

Any help much appreciated.

Thank you a lot!

 

  • Hi Anonymous ,

     

    Is it this you are looking for?

    I solved it in two steps with two different calculated columns.

    First, we need to create an index column so we can iterate through each row and calculate the difference between their respective dates. Since we have rows which have the same date, we do need to have a second column deciding which one to rank first or second. In your case, I chose to_stage_id, so the ranking looks like this:

     

    index = 
    VAR MaxToStageID = MAX( 'Table'[to_stage_id] )
    VAR result =
        RANKX (
            ALL ( 'Table' ),
            'Table'[activity_created] * MaxToStageID + 'Table'[to_stage_id],
            ,
            ASC,
            DENSE
        )
    RETURN
        result

     

    The guys from sqlbi have done a blog about this topic on how to rank / create indexes on multiple columns.

     

    After that, I created another calculated column which uses the index to iterate through the table calculating the date difference:

     

     

    tomstest = 
    VAR Index =  ('Table'[index])
    VAR PreviousIndex = ('Table'[index] - 1 )
    VAR result = 
            DATEDIFF( 
                CALCULATE ( 
                    VALUES ( 'Table'[activity_created] ),
                    FILTER ( ALL ('Table'), 'Table'[index] = PreviousIndex )  
                ),
                CALCULATE ( 
                    VALUES ( 'Table'[activity_created] ),
                    FILTER ( ALL ('Table'), 'Table'[index] = Index )  
                ),
                DAY
            )
    RETURN result

     

    Hope this helps!


    /Tom

    https://www.tackytech.blog

    https://www.instagram.com/tackytechtom

7 Replies

  • Anonymous's avatar
    Anonymous
    Not applicable

    I have tried this formula in a calculated column, it is near what i what but not exactly.

     

    Test Column =
    VAR temp =
    TOPN (
    1,
    FILTER (
    'table',
    'table''[candidate_id] = EARLIER ( 'table''[candidate_id])
    && 'table''[activity_created] < EARLIER ( 'table''[activity_created])
    ),
    'table''[activity_created], DESC
    )
    RETURN
    DATEDIFF ( MINX ( temp, 'table''[activity_created] ), 'table''[activity_created], DAY )
     
    This is the result I got

    Reviewing -1st Interview (14 May) - 1 day not 3.

    and the total of  48 is not not correct. there are 35 days (11 may-14 june).

    Can you also show me how can I calculate the correct total days from start date till the end date?

    • tackytechtom's avatar
      tackytechtom
      Most Valuable Professional

      Hi Anonymous ,

       

      Is it this you are looking for?

      I solved it in two steps with two different calculated columns.

      First, we need to create an index column so we can iterate through each row and calculate the difference between their respective dates. Since we have rows which have the same date, we do need to have a second column deciding which one to rank first or second. In your case, I chose to_stage_id, so the ranking looks like this:

       

      index = 
      VAR MaxToStageID = MAX( 'Table'[to_stage_id] )
      VAR result =
          RANKX (
              ALL ( 'Table' ),
              'Table'[activity_created] * MaxToStageID + 'Table'[to_stage_id],
              ,
              ASC,
              DENSE
          )
      RETURN
          result

       

      The guys from sqlbi have done a blog about this topic on how to rank / create indexes on multiple columns.

       

      After that, I created another calculated column which uses the index to iterate through the table calculating the date difference:

       

       

      tomstest = 
      VAR Index =  ('Table'[index])
      VAR PreviousIndex = ('Table'[index] - 1 )
      VAR result = 
              DATEDIFF( 
                  CALCULATE ( 
                      VALUES ( 'Table'[activity_created] ),
                      FILTER ( ALL ('Table'), 'Table'[index] = PreviousIndex )  
                  ),
                  CALCULATE ( 
                      VALUES ( 'Table'[activity_created] ),
                      FILTER ( ALL ('Table'), 'Table'[index] = Index )  
                  ),
                  DAY
              )
      RETURN result

       

      Hope this helps!


      /Tom

      https://www.tackytech.blog

      https://www.instagram.com/tackytechtom

      • Cortana's avatar
        Cortana
        Helper IV

        Hello there!

        I was looking for something similar. But found this. 

        My question is my interest is to calculate how long the opportunities are staying in each stage. So in your case, when the opportunity entered stage 2, the date was the 14th of May, and when it entered stage 3 the date was the 19th of May. So the duration for stage 2 for this opportunity should be 5. But in your case, it is showing 0 days. That is the result of the previous stage. 

        Did I get it wrong?

  • Anonymous's avatar
    Anonymous
    Not applicable

    Hi Tom, Thank you for the effort and your solution .

    Yes, what you have achieved will work for me as a final goal, but when I replicate your solution in my data I get 0 as a result. 

    I think the problem lies in the column  to_stage_id. The Id's are maybe not a good idea for sorting.

     

    I have also tried to create the Index column with this logic, but its not giving me the desired result.

    Index =
    RANKX('Job Application FT', 'Job Application FT'[activity_created].[Date], , ASC, Dense)
     
    Any other idea what could be the problem?
  • v-luwang-msft's avatar
    v-luwang-msft
    Community Support

    Hi Anonymous  Anonymous ,

    You could also  create index through Power Query Editor:

     

    And you could download pbix file to check with your file.

     

     

    Best Regards

    Lucien