Calculate difference in minutes between rows (example included)

[dnewton]

Hi,

I'm trying to create a column or measure to calculate the amount of time agents have spent in "withdrawn".

In the below example you can see my data -

Starting from the bottom of the table you can see the agent has gone into state e_state "Agent Withdrawn" at e_UTC "10/12/2017 06:18:55" and then went into e_state "Agent Assigned" at e_UTC 10/12/2017 06:30:35.

How can I automatically calculate how much time was spent in "Agent Withdrawn"? So the time between Agent Withdrawn and Agent Assigned i.e. 10/12/2017 06:18:55 to 10/12/2017 06:30:35 = 11 minutes and 40 seconds.

[Greg_Deckler · Accepted solution]

The calculation goes something along these lines (I developed this is for a different use case but same problem):

c_Duration = 

VAR next = MINX(
FILTER(Table,
Table[AgentName]=EARLIER(Table[AgentName]) && Table[Time]>EARLIER(Table[Time])
),
Table[Time])

RETURN IF(ISBLANK(next),DATEDIFF([Time],NOW(),SECOND),DATEDIFF([Time],next,SECOND))

This is a calculated column in your table that will return the Duration in seconds between one row and the next row. So basically what is happening is that you are creating a table that filters down to the same Agent as the current row and all Times that are after the current row. Get the earliest date of that table (MINX). If there is a blank, you are at the top row in the dataset so get the Duration between that time an NOW(). 

Please note that this is using MINX so using this on a dataset that contains like 11 million rows is a bad idea but it works great for smaller tables.

[dnewton]

This is amazing, I have been trying on and off for days to get a calculation and this seems to be working!

I suppose the next question is I now have a column full of total seconds, I need to get that to display in mins.

I guess [c_Duration]/60 would be too easy!

[Greg_Deckler]

Yep, that's how I would do it. Or DIVIDE([c_Duration],60) but same thing! Or, just change SECOND in your DATEDIFFS to MINUTE.