Correlation between same column, different items for time periods

[Anonymous]

Hi All,

I have a data table consisting of call data. Each different call shares a reference "Session ID" and each time something happens within the call this is logged with a time stamp. I am wanting to be able to get the length of each call based on this info. I can then use this for the average time taken to complete actions etc.

Here is part of my data table.

Logtimestamp	Session ID
14/12/2018 11:53:19	ABC-001
14/12/2018 11:53:25	ABC-001
14/12/2018 11:53:40	ABC-001
14/12/2018 11:54:10	ABC-001
14/12/2018 11:54:25	ABC-001
14/12/2018 11:54:37	ABC-001
14/12/2018 11:54:50	ABC-001
14/12/2018 11:54:55	ABC-001
14/12/2018 11:55:15	ABC-001
14/12/2018 11:53:50	ABC-002
14/12/2018 11:54:10	ABC-002
14/12/2018 11:56:05	ABC-002

and I want to end up with

Session ID	Start	End	Durration (Secs)
ABC-001	14/12/2018 11:53:19	14/12/2018 11:55:15	56
ABC-002	14/12/2018 11:53:50	14/12/2018 11:56:05	15

any idea how I do this?

the timestamps might not be in any particular order either.

[Anonymous · Accepted solution]

Thanks Greg,

I've managed to find a solution.

[Greg_Deckler]

See my article on Mean Time Before Failure (MTBF) which uses EARLIER: http://community.powerbi.com/t5/Community-Blog/Mean-Time-Between-Failure-MTBF-and-Power-BI/ba-p/3395...

[Anonymous]

Hi,

thanks for your help, however I'm not sure how this gets me what I'm after. I've tried to keep up with your article but I am struggling to apply it to my scenario.

Ideally I would like a new table which shows the earliest timestamp of a call ID in 1 column and the latest timestamp of it in another column.